Module Handbook

for the

Master Programme “Computer Science”

at

Rheinischen Friedrich-Wilhelms-Universität Bonn

revised version: March 28, 2026

The curriculum of the master programme is divided into four sub-curricula, each corresponding to one of
the four main areas of competence in research of the Bonn Institute of Computer Science:

1. Algorithmics
2. Graphics, Vision, Audio
3. Information and Communication Management
4. Intelligent Systems

Module numbers MA-INF ASXY have been assigned according to the following key: vergeben:

• A = number of the area of competence

• S = semester within the master curriculum
• XY = sequential number within the semester and the respective area of competence (two digits)

According to the curriculum, all modules ought to be taken between the first and the third semester. The
fourth semester is reserved for preparing the master thesis.

Contents

1 Algorithmics 2

2 Graphics, Vision, Audio 43

3 Information and Communication Management 84

4 Intelligent Systems 111

5 Master Thesis 168

Master Computer Science — Universität Bonn 2

1 Algorithmics

MA-INF 1102 L4E2 9 CP Combinatorial Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

MA-INF 1103 L4E2 9 CP Cryptography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
MA-INF 1105 L2E2 6 CP Algorithms for Data Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
MA-INF 1107 L2E2 6 CP Foundations of Quantum Computing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
MA-INF 1108 L2E2 6 CP Introduction to High Performance Computing: Architecture Features and
Practical Parallel Programming . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
MA-INF 1110 L2E2 6 CP Elliptic Curves and their Applications in Cryptography . . . . . . . . . . . . . . . . . . . 9
MA-INF 1201 L4E2 9 CP Approximation Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
MA-INF 1202 L4E2 9 CP Chip Design . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
MA-INF 1203 L4E2 9 CP Discrete and Computational Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
MA-INF 1205 6 CP Graduate Seminar Discrete Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
MA-INF 1206 Sem2 4 CP Seminar Randomized and Approximation Algorithms . . . . . . . . . . . . . . . . . . . . 15
MA-INF 1209 Sem2 4 CP Seminar Advanced Topics in Cryptography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
MA-INF 1213 L4E2 9 CP Randomized Algorithms and Probabilistic Analysis . . . . . . . . . . . . . . . . . . . . . . 17
MA-INF 1218 L4E2 9 CP Algorithms and Uncertainty . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
MA-INF 1219 Sem2 4 CP Seminar Algorithmic Game Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
MA-INF 1221 Lab4 9 CP Lab Computational Analytics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
MA-INF 1222 Lab4 9 CP Lab High Performance Optimization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
MA-INF 1223 L4E2 9 CP Privacy Enhancing Technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
MA-INF 1224 L2E2 6 CP Quantum Computing Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
MA-INF 1225 Lab4 9 CP Lab Exploring HPC technologies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
MA-INF 1226 L2E2 6 CP Applications of Computational Topology in Information Theory . . . . . . . . . . 29
MA-INF 1228 Sem2 6 CP Graduate Seminar on Algorithms and Optimization . . . . . . . . . . . . . . . . . . . . . . 30
MA-INF 1229 Lab4 9 CP Lab Advanced Algorithm Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
MA-INF 1230 L2E2 6 CP Algorithmic Knot Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
MA-INF 1301 L4E2 9 CP Algorithmic Game Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
MA-INF 1307 Sem2 4 CP Seminar Advanced Algorithms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
MA-INF 1309 Lab4 9 CP Lab Efficient Algorithms: Design, Analysis and Implementation . . . . . . . . . . 36
MA-INF 1314 L4E2 9 CP Online Motion Planning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
MA-INF 1315 Lab4 9 CP Lab Computational Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39
MA-INF 1316 Lab4 9 CP Lab Cryptography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
MA-INF 1322 Sem2 4 CP Seminar Focus Topics in High Performance Computing . . . . . . . . . . . . . . . . . . 41
MA-INF 1323 L4E2 9 CP Computational Topology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
Master Computer Science — Universität Bonn 3

MA-INF 1102 Combinatorial Optimization

Workload Credit points Duration Frequency

270 h 9 CP 1 semester at least every year

Module coordinator Lecturer(s)

Prof. Dr. Jens Vygen All lecturers of Discrete Mathematics

Programme Mode Semester

M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

Advanced knowledge of combinatorial optimization. Modelling and development of solution strategies for
combinatorial optimization problems

Learning goals: soft skills

Mathematical modelling of practical problems, abstract thinking, presentation of solutions to exercises

Contents
Matchings, b-matchings and T-joins, optimization over matroids, submodular function minimization, travelling
salesman problem, polyhedral combinatorics, NP-hard problems

Prerequisites
none
Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam
Ungraded coursework (required for admission to the exam)
The completion of regularly provided exercise sheets. The work can be done individually or in groups of two
students. A total of 50% of the points must be achieved. Each student must present a solution to an exercise in the
exercise sessions twice.
Literature
• B. Korte, J. Vygen: Combinatorial Optimization: Theory and Algorithms. Springer, 6th edition, 2018
• A. Schrijver: Combinatorial Optimization: Polyhedra and Efficiency. Springer, 2003
• W. Cook, W. Cunningham, W. Pulleyblank, A. Schrijver: Combinatorial Optimization. Wiley, 1997
• A. Frank: Connections in Combinatorial Optimization. Oxford University Press, 2011
Master Computer Science — Universität Bonn 4

MA-INF 1103 Cryptography

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Dr. Michael Nüsken Dr. Michael Nüsken
Programme Mode Semester
M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

Understanding of security concerns and measures, and of the interplay between computing power and security
requirements. Mastery of the basic techniques for cryptosystems and cryptanalysis, including modelling security and
reducing security to basic assumptions.

Learning goals: soft skills

Competences: Ability to assess, present and explain schemes and their use in applications, orally and written.
Critical assessment of applications in terms of security, social and ethical context and more.

Contents
Basic private-key and public-key cryptosystems: AES, RSA, group-based. Security reductions. Key exchange,
cryptographic hash functions, signatures, identification; factoring integers and discrete logarithms; lower bounds in
structured models.
Prerequisites
Recommended:
Basics in elementary number theory, groups and complexity theory -in particular, reductions- are helpful.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. Each student must present a solution to an exercise in the
exercise sessions twice.
Literature
• Jonathan Katz & Yehuda Lindell (2015/2008). Introduction to Modern Cryptography, CRC Press.
• Course notes
Master Computer Science — Universität Bonn 5

MA-INF 1105 Algorithms for Data Analysis

Workload Credit points Duration Frequency

180 h 6 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Petra Mutzel Prof. Dr. Petra Mutzel
Programme Mode Semester
M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

Ability to independently design and analyze efficient algorithms and data structures, in particular using methods
and techniques of modern algorithmics with respect to big data and/or analytics tasks;

Learning goals: soft skills

Presentation of solutions and methods; critical discussion of applied methods and techniques clearly and in
accordance with academic standards; ability to analyze problems theoretically and to find efficient as well as
practical solutions; to examine one’s solutions and results critically; to classify new problems into the state-of-the-art
of the respective area;

Contents
Advanced algorithmic techniques and data structures relevant to analytic tasks for big data, i.e., algorithms for
efficiently computing centrality indices for networks, theoretical and practical approaches to graph similarity, parallel
algorithms, external data structures, and streaming algorithms.

Prerequisites
Recommended:
Essential is knowledge of:
• fundamental algorithms and algorithmic paradigms (e.g., graph algorithms, greedy algorithms, divide and conquer,
dynamic programming), data structures (e.g., balanced search trees, hash tables)
• mathematical foundations of algorithm analysis (e.g., Big O notation, recurrence relations, proof techniques,
running-time analysis)
• computational complexity (e.g., NP-hardness, reductions)

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam (30 minutes)

Ungraded coursework (required for admission to the exam)

It is necessary to pass one of two written tests during the semester. Moreover, the completion of regularly provided
exercise sheets is required. The work can be done in groups. A total of 50% of the points must be achieved. Each
student must present a solution to an exercise in the exercise sessions twice. At the beginning of each exercise
session, all participants mark on a list which (sub)exercises they have completed successfully and for which they wish
to receive credit. The tutor then selects, for each (sub)exercise, one participant to present it. For more complex
exercises, a written solution is required, which can be uploaded during the presentation.
Master Computer Science — Universität Bonn 6

MA-INF 1107 Foundations of Quantum Computing

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every 2 years

Module coordinator Lecturer(s)

Prof. Dr.-Ing. Christian Bauckhage Prof. Dr.-Ing. Christian Bauckhage

Programme Mode Semester

M. Sc. Computer Science Optional 1. or 3.

Learning goals: technical skills

Upon successful completion of this module, students should be able to describe fundamental concepts and techniques
(qubits, quantum registers, quantum gates, quantum circuits) in quantum computing. Students will be equipped
with specific, quantum computing related programming know-how; based on knowledge and skills acquired, students
should be able to
• devise quantum computing algorithms for basic computational tasks
• run these algorithms on (simulated) quantum computers

Learning goals: soft skills

In the exercises, students will have the opportunity to put their knowledge into practice, since they will realize small
projects on computing with quantum gates and their solutions using quantum inspired methods or genuine quantum
methods. This requires teamwork; upon successful completion of the module, students should be able to
• draft and implement basic quantum computing algorithms
• apply quantum computing (simulations) to test these algorithms
• prepare and give oral presentations about their work in front of an audience

Contents
Boolean algebras and Boolean lattices; cellular automata; classical digital computing; classical reversible computing;
mathematical foundations of quantum computing (complex vector spaces, tensor products, unitary operators,
Hermitian operators, qubits, superposition, entanglement); quantum gate computing; quantum circuits

Prerequisites
Recommended:
Good working knowledge of theory and practice of linear algebra

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam
Ungraded coursework (required for admission to the exam)
The completion of regularly provided exercise sheets. The work can be done in groups of up to five students. A total
of 50% of the points must be achieved. Each student must present a solution to an exercise in the exercise sessions
once.
Forms of media
• lecture slides / lecture notes are made available online
• notebooks with programming examples are made available online
Master Computer Science — Universität Bonn 7

Literature
• L. Susskind, A. Friedman, “Quantum Mechanics: The Theoretical Minimum”, Penguin, 2015
• M.A. Nielsen, I.L Chuang, “Quantum Computation and Quantum Information”, Cambridge University Press, 10th
Anniversary edition, 2010
• P. Wittek, “Quantum Machine Learning”, Academic Press, 2016
• M. Schuld, F. Petruccione, “Machine Learning with Quantum Computers”, Springer, 2nd edition, 2021
• S. Ganguly, “Quantum Machine Learning: An Applied Approach”, Apress, 2021
Master Computer Science — Universität Bonn 8

MA-INF 1108 Introduction to High Performance Computing: Architecture

Features and Practical Parallel Programming

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Estela Suarez Prof. Dr. Estela Suarez
Programme Mode Semester
M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

Understanding principles of computer architecture of modern HPC systems at component (processor, accelerators)
and system level (system architecture, network, memory hierarchy) and their implication for application
programming. Ability to program parallel computers, employing multi-core and multi-node features. Programming
CPU and GPUs. Understanding the quality of performance and scaling behaviour, and applying the measures
needed to improve them.

Learning goals: soft skills

Critical assessment of hardware and applications in terms of performance and efficiency.

Contents
• Computer architectures, system components (CPU, memory, network) and their interrelation.
• Software environment
• Access to HPC compute resources at the Jülich Supercomputing Centre
• Practical use of parallel programming paradigms (MPI, OpenMP, CUDA)
• Performance of applications and scaling behavior, understanding and strategies for improvement
• Current challenges in HPC

Prerequisites
Required:
MA-INF 1108 replaces MA-INF 1106 and cannot be taken after completing MA-INF 1106.
Recommended:
Knowledge of a modern programming language (ideally C/C++ and Python) is required.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (90 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. A total of 50% of the points must be achieved. Each student
must present a solution to an exercise in the exercise sessions once.

Forms of media
Laptop and projector

Literature
• John L. Hennessy, David A. Patterson: Computer Architecture - A Quantitative Approach. Morgan Kaufmann
Publishers, 2012
• David A. Patterson, John L. Hennessy: Computer Organization and Design - The Hardware / Software Interface.
Morgan Kaufmann Publishers, 2013
• Message Passing Interface Forum: MPI: A Message-Passing Interface Standard, Version 3.1
• OpenMP Application Programming Interface, Version 4.5, November 2015
Master Computer Science — Universität Bonn 9

MA-INF 1110 Elliptic Curves and their Applications in Cryptography

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Michael Meier Dr. Gerhard Schabhüser
Programme Mode Semester
M. Sc. Computer Science Optional 1-3.

Learning goals: technical skills

After completion of the modul students will understand the theory of elliptic Curves and their applications in
cryptography. The students will learn, why elliptic curves have useful properties in Cryptography (at least as long as
there are no powerful Quantum-Computers in place).
The relevant mathematical background will be developed and specific applications like
ECC-Pseudo-Random-Number-Generators, ECC-Public-KC-Protocols and-Algorithms or Algorithms used in
Cryptoanalysis will be presented.
In addition to the mathematical perspective the implementation-perspective especially concerning secure
implementation will be addressed.

Learning goals: soft skills

Students will get the competence of understanding the mathematical background of elliptic curves, they will be
sensible for the need of secure implementation of cryptographic algorithms and -protocols. They will be aware, that
understanding the mathematics of cryptography has the capability to support secure implementation in
cryptography their knowledge for secure implementation.
Master Computer Science — Universität Bonn 10

Contents
Algebraic Background
• algebraic Structures
• Introduction to Number theory
• Finite Fields
Basic Arithmetric Algorithms.
• Exponentiation.
• Integer arithmetic.
• Finite fields arithmetic.
Geometry und arithmetic of algebraic curves.
• Affine space.
• Projective space.
Elliptic Curves.
• Arithmetic.
• Points of finite order.
• Group of rational points.
Cubic curves over finite fields.
• Arithemetic.
• Rational points.
• A theorem of Gauß.
• Points of finite order (revisor).
Elliptic curve cryptography.
• Factorisation of integers.
• Discrete logarithm.
• Diffie-Hellman-protocol.
• El-Gamal-signatures.
• EC-DSA.
Implementing ECC-arithmetic.
• Timing attacks.
• Power consumption attacks.
• Secure implementions using Mathematics.

Prerequisites
Recommended:
Recommended : linear algebra, basic knowledge in algebra.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam (30 min).

Ungraded coursework (required for admission to the exam)

Participation in one written achievement test. At least 50% of the points much be achieved.

Literature
• Henri Cohen, Gerhard Frey et. al. (2005). Handbook of Elliptic and hyperelliptic Curve Cryptography.
• Joseph Silverman and John Tate. Rational Points on Elliptic Curves.
• Joseph Silverman. The Arithmetic of Elliptic Curves. 2nd Edition.
Master Computer Science — Universität Bonn 11

MA-INF 1201 Approximation Algorithms

Workload Credit points Duration Frequency

270 h 9 CP 1 semester at least every year

Module coordinator Lecturer(s)

Prof. Dr. Jens Vygen All lecturers of Discrete Mathematics, Senior Prof. Dr. Marek Karpinski

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Introduction to design and analysis of most important approximation algorithms for NP-hard combinatorial
optimization problems, and various techniques for proving lower and upper bounds, probabilistic methods and
applications

Learning goals: soft skills

Presentation of solutions and methods, critical discussion of applied methods and techniques

Contents
Approximation Algorithms and Approximation Schemes. Design and Analysis of Approximation algorithms for
selected NP-hard problems, like Set-Cover, and Vertex-Cover problems, MAXSAT, TSP, Knapsack, Bin Packing,
Network Design, Facility Location. Introduction to various approximation techniques (like Greedy, LP-Rounding,
Primal-Dual, Local Search, randomized techniques and Sampling, and MCMC-Methods), and their applications.
Analysis of approximation hardness and PCP-Systems.

Prerequisites
Recommended:
Introductory knowledge of foundations of algorithms and complexity theory is essential.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam
Ungraded coursework (required for admission to the exam)
The completion of regularly provided exercise sheets. The work can be done in groups of up to two students. A total
of 50% of the points must be achieved. Each student must present a solution to an exercise in the exercise sessions
twice.
Literature
• S. Arora, C. Lund: Hardness of Approximations. In: Approximation Algorithms for NP-Hard Problems (D. S.
Hochbaum, ed.), PWS, 1996
• M. Karpinski: Randomisierte und approximative Algorithmen für harte Berechnungsprobleme, Lecture Notes (5th
edition), Universität Bonn, 2007
• B. Korte, J. Vygen: Combinatorial Optimization: Theory and Algorithms (6th edition), Springer, 2018
• V. V. Vazirani: Approximation Algorithms, Springer, 2001
• D. P. Williamson, D. B. Shmoys: The Design of Approximation Algorithms, Cambridge University Press, 2011
Master Computer Science — Universität Bonn 12

MA-INF 1202 Chip Design

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Jens Vygen All lecturers of Discrete Mathematics

Programme Mode Semester

M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

Knowledge of the central problems and algorithms in chip design. Competence to develop and apply algorithms for
solving real-world problems, also with respect to technical constraints. Techniques to develop and implement efficient
algorithms for very large instances.

Learning goals: soft skills

Mathematical modelling of problems occurring in chip design, development of efficient algorithms, abstract thinking,
presentation of solutions to exercises

Contents
Problem formulation and design flow for chip design, logic synthesis, placement, routing, timing analysis and
optimization

Prerequisites
none
Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam
Ungraded coursework (required for admission to the exam)
The completion of regularly provided exercise sheets. The work can be done in groups of up to two students. A total
of 50% of the points must be achieved. Each student must present a solution to an exercise in the exercise sessions
twice.
Literature
• C.J. Alpert, D.P. Mehta, S.S. Sapatnekar: The Handbook of Algorithms for VLSI Physical Design Automation.
CRC Press, New York, 2008.
• S. Held, B. Korte, D. Rautenbach, J. Vygen: Combinatorial optimization in VLSI design. In: "Combinatorial
Optimization: Methods and Applications" (V. Chvátal, ed.), IOS Press, Amsterdam 2011, pp. 33-96
• S. Held, J. Vygen: Chip Design. Lecture Notes (distributed during the course)
• L. Lavagno, I.L. Markov, G. Martin, and L.K. Scheffer, eds.: Electronic Design Automation for IC Implementation,
Circuit Design, and Process Technology. CRC Press, 2nd edition, 2016
Master Computer Science — Universität Bonn 13

MA-INF 1203 Discrete and Computational Geometry

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Anne Driemel Prof. Dr. Anne Driemel, PD Dr. Elmar Langetepe, Dr. Herman Haverkort

Programme Mode Semester

M. Sc. Computer Science Optional 1-4.

Learning goals: technical skills

Knowledge of fundamental theorems and concepts in the area of discrete and computational geometry; design and
analysis of geometric algorithms; combinatorial analysis of the complexity of geometric configurations; to apply this
knowledge autonomously in solving new problems.

Learning goals: soft skills

Social competence (communication, presenting one’s own solutions, goal-oriented discussions in teams), methodical
competence (analysis, abstraction, proofs), individual competence (commitment and willingness to learn, creativity,
endurance).

Contents
The course offers an overview of fundamental theorems and techniques in the area of Computational Geometry. The
main focus lies on the discussion and analysis of geometric algorithms and data structures, as well as the complexity
of geometric structures, such as Voronoi-diagrams, Delaunay triangulations, polytopes, and arrangements in higher
dimensions. In terms of algorithmic techniques, we focus on the randomized incremental construction of these
geometric structures, as well as data structures for point location and approximate near-neighbors. We will also
discuss geometric concepts, such as center points and halfspace depth, set systems, their VC-dimension and
epsilon-nets and show their application to fundamental problems, such as the art gallery problem.

Prerequisites
Recommended:
BA-INF 114 – Grundlagen der algorithmischen Geometrie

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam
Ungraded coursework (required for admission to the exam)
The completion of regularly provided exercise sheets. The work can be done in groups of up to two students. A total
of 50% of the points must be achieved.

Literature
• Jiri Matousek. Lectures on Discrete Geometry. Springer Graduate Texts in Mathematics. ISBN 0-387-95374-4.
• Mark de Berg, Otfried Cheong, Marc van Kreveld, and Mark Overmars. Computational Geometry — Algorithms
and Applications (Third Edition). Springer. ISBN 978-3-540-77973-5.
Master Computer Science — Universität Bonn 14

MA-INF 1205 Graduate Seminar Discrete Optimization

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Jens Vygen All lecturers of Discrete Mathematics

Programme Mode Semester

M. Sc. Computer Science Optional 2.

Learning goals: technical skills

Competence to understand new research results based on original literature, to put such results in a broader context
and present such results and relations.

Learning goals: soft skills

Ability to read and understand research papers, abstract thinking, presentation of mathematical results in a talk

Contents
A current research topic in discrete optimization will be chosen each semester and discussed based on original
literature.
Prerequisites
Recommended:
MA-INF 1102 – Combinatorial Optimization

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 4 60 T / 120 S 6

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
The topics and the relevant literature will be announced towards the end of the previous semester.
Master Computer Science — Universität Bonn 15

MA-INF 1206 Seminar Randomized and Approximation Algorithms

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Heiko Röglin Prof. Dr. Anne Driemel, Prof. Dr. Thomas Kesselheim,
Prof. Dr. Heiko Röglin, PD Dr. Elmar Langetepe, Dr. Herman Haverkort,
Senior Prof. Dr. Marek Karpinski

Programme Mode Semester

M. Sc. Computer Science Optional 2.

Learning goals: technical skills

Ability to perform individual literature search, critical reading, understanding, and clear presentation.

Learning goals: soft skills

Presentation of solutions and methods, critical discussion of applied methods and techniques

Contents
Current topics in design and analysis of randomized and approximation algorithms based on lastest research
literature
Prerequisites
none
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
The relevant literature will be announced in time.
Master Computer Science — Universität Bonn 16

MA-INF 1209 Seminar Advanced Topics in Cryptography

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every semester

Module coordinator Lecturer(s)

Dr. Michael Nüsken Dr. Michael Nüsken
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Enhanced and in-depth knowledge in specialized topics in the area of cryptography.

Learning goals: soft skills

Acquire the competence to independently search for and study state-of-the-art scientific literature in depth, read
critically, identify the most relevant content, and assess research results in the context of cryptography; to discuss
research results with a knowledgeable scientific audience; to present prior work by others in writing and in
presentations with visual media at academic standards, well-structured and didactically effective, and motivate the
audience to participate; to formulate and accept criticism; to manage one’s time with relatively open assignments
and long-ranging deadlines.

Contents
We discuss cutting-edge papers from current cryptographic research literature.

Prerequisites
Recommended:
Basic knowledge in cryptography is highly recommended, eg. by MA-INF 1103 – Cryptography.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
Current cryptographic literature.
Master Computer Science — Universität Bonn 17

MA-INF 1213 Randomized Algorithms and Probabilistic Analysis

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Heiko Röglin Prof. Dr. Heiko Röglin

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 4.

Learning goals: technical skills

Understanding of models and techniques for the probabilistic analysis of algorithms as well as for the design and
analysis of randomized algorithms

Learning goals: soft skills

Oral and written presentation of solutions and methods, abstract thinking

Contents
Design and analysis of randomized algorithms
• complexity classes
• Markov chains and random walks
• tail inequalities
• probabilistic method
smoothed and average-case analysis
• simplex algorithm
• local search algorithms
• clustering algorithms
• combinatorial optimization problems
• multi-objective optimization

Prerequisites
none
Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam
Ungraded coursework (required for admission to the exam)
The completion of regularly provided exercise sheets. The work can be done in groups of up to three students. A
total of 25% of the points must be achieved.

Literature
• lecture notes
• research articles
• Motwani, Raghavan, Randomized Algorithms, Cambridge University Press, 1995
• Mitzenmacher, Upfal, Probability and Computing, Cambridge University Press, 2nd edition, 2017
Master Computer Science — Universität Bonn 18

MA-INF 1218 Algorithms and Uncertainty

Workload Credit points Duration Frequency

270 h 9 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Thomas Kesselheim Prof. Dr. Thomas Kesselheim
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Understanding approaches for modeling uncertainty in algorithmic theory. Designing and analyzing algorithms with
performance guarantees in the context of uncertainty.

Learning goals: soft skills

Oral and written presentation of solutions and methods

Contents
• Advanced Online Algorithms
• Markov Decisions Processes
• Stochastic and Robust Optimization
• Online Learning Algorithms and Online Convex Optimization

Prerequisites
Recommended:
Solid background in algorithms, calculus, and probability theory. Specialized knowledge about certain algorithms is
not necessary.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam
Ungraded coursework (required for admission to the exam)
Each student must present a solution to an exercise in the exercise sessions once.

Literature
lecture notes, research articles
Master Computer Science — Universität Bonn 19

MA-INF 1219 Seminar Algorithmic Game Theory

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Thomas Kesselheim Prof. Dr. Thomas Kesselheim
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Ability to understand new research results presented in original scientific papers.

Learning goals: soft skills

Ability to perform individual literature search, critical reading, and clear didactic presentation

Contents
Advanced topics in Algorithmic Game Theory and Algorithmic Mechanism Design based on current conference and
journal papers

Prerequisites
none
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 20

MA-INF 1221 Lab Computational Analytics

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Petra Mutzel Prof. Dr. Petra Mutzel
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Ability to independently design, theoretically analyze, implement, and experimentally evaluate algorithms and
efficient data structures for computational analytics problems; gain experience with software development
techniques, tools and standards and the scientifically clean documentation of the students own work (including the
written report and software).

Learning goals: soft skills

• Knowledge of scientific approach to problem solving;
• ability to scientifically present solutions and methods;
• critical discussion of applied methods and techniques clearly and in accordance with academic standards;
• ability to analyze problems theoretically and to find efficient as well as practical solutions;
• to examine one’s solutions and results critically;
• to classify new problems into the state-of-the-art of the respective area.

Contents
We will design efficient exact and approximate algorithms and data structures for computational analytics problems.
We study a (set of) selected combinatorial optimization problem(s) with the goal to design new algorithmic
approaches. Often, we focus on solving (graph) problems for selected applications (e.g., in cartography, geodesy,
neurosciences, chemistry, or others). Typically, we start with a literature search on State-of-the-Art approaches;
based on that, we adapt selected approaches
to our studied problem(s) or we design new approaches. We then theoretically analyze and implement our
adapted/new algorithms. This is followed by an extensive experimental evaluation including a discussion of the
results on benchmark instances. Often, the analysis triggers improvements of the algorithms. This is also called the
Algorithm Engineering cycle.

Prerequisites
Recommended:
Essential are knowledge of:
• fundamental algorithms and algorithmic paradigms (e.g., graph algorithms, greedy algorithms, divide and conquer,
dynamic programming), data structures (e.g., balanced search trees, hash tables)
• mathematical foundations of algorithm analysis (e.g., Big O notation, recurrence relations, proof techniques,
running-time analysis)
• computational complexity (e.g., NP-hardness, reductions).
It is recommended to first complete at least one of the following modules:
• MA-INF 1105 Algorithms for Data Analysis
• MA-INF 1201 Approximation Algorithms
• MA-INF 1203 Discrete and Computational Geometry
• MA-INF 1213 Randomized Algorithms and Probabilistic Analysis
• MA-INF 1218 Algorithms and Uncertainty
• MA-INF 1301 Algorithmic Game Theory
• MA-INF 4112 Algorithms for Data Science

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report
Master Computer Science — Universität Bonn 21

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
The relevant literature will be announced in time.
Master Computer Science — Universität Bonn 22

MA-INF 1222 Lab High Performance Optimization

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Petra Mutzel Prof. Dr. Petra Mutzel, Gayatri Manda

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

• Ability to independently design, theoretically analyze, implement, and experimentally evaluate algorithms and
efficient data structures for computational analytics problems;
• understanding and using parallel programming paradigms and high-level programming languages;
• using performance analysis tools, understanding performance bottlenecks and measures to improve them;
• acquisition of knowledge about software development and standards;
• gain experience with the documentation of the students own work (including the written report and software);

Learning goals: soft skills

• Knowledge of scientific approach to problem solving;
• ability to scientifically present solutions and methods;
• critical discussion of applied methods and techniques clearly and in accordance with academic standards;
• ability to analyze problems theoretically and to find efficient as well as practical solutions;
• to examine one’s solutions and results critically;
• to classify new problems into the state-of-the-art of the respective area;

Contents
We will design efficient exact and approximate algorithms and data structures for optimization problems on big data
with the focus of using high performance computing (HPC) systems (like, e.g. the HPC clusters Marvin or Bender).
We study a (set of) selected optimization problem(s) with the goal to design new parallel algorithms that scale well
on HPC systems. Often, we focus on solving (graph) problems for selected applications (e.g., physics, chemistry,
neurosciences, geodesy, or others).
Typically, we start with an introduction into parallel algorithms and an introduction into the relevant API for
developing parallel programs. A literature search yields State-of-the-Art techniques; based on that, we adapt selected
approaches to our studied problem(s) or we design new approaches with the goal that they scale well on HPC
systems. We then theoretically analyze and implement our adapted/new parallel algorithms using parallel
programming paradigms and high-level programming languages. This is followed by an extensive experimental
evaluation using performance analysis tools and understanding performance bottlenecks. Often, this triggers
improvements of the parallel algorithms and/or the implementation.

Prerequisites
Recommended:
Essential are knowledge of:
• fundamental algorithms and algorithmic paradigms (e.g., graph algorithms, greedy algorithms, divide and conquer,
dynamic programming), data structures (e.g., balanced search trees, hash tables)
• mathematical foundations of algorithm analysis (e.g., Big O notation, recurrence relations, proof techniques,
running-time analysis)
• computational complexity (e.g., NP-hardness, reductions)
It is recommended to complete at least one the following modules first:
• MA-INF 1105 Algorithms for Data Analysis
• MA-INF 1108 Introduction to High Performance Computing: Architecture Features and Practical Parallel
Programming
• MA-INF 1201 Approximation Algorithms
• MA-INF 1203 Discrete and Computational Geometry
• MA-INF 1213 Randomized Algorithms and Probabilistic Analysis
• MA-INF 1218 Algorithms and Uncertainty
• MA-INF 1301 Algorithmic Game Theory
Master Computer Science — Universität Bonn 23

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
The relevant literature will be announced in time.
Master Computer Science — Universität Bonn 24

MA-INF 1223 Privacy Enhancing Technologies

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Dr. Michael Nüsken Dr. Michael Nüsken
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Knowledge: Cryptographic schemes for enhancing privacy, underlying security notions, applications and restrictions.
Skills: Secure application of sophisticated cryptographic schemes. Evaluation of their correctness, efficiency and
security in an application setting.

Learning goals: soft skills

Competences: Ability to assess, present and explain schemes and their use in applications, orally and written.
Critical assessment of applications in terms of security, social and ethical context and more.

Contents
With more and more data available a clear separation of sensitive data is necessary and needs to be protected. Some
of that data must stay within strict environments, for examples hospitals must store certain highly sensitive medical
information about patients but they are not allowed to store it outside its own facilities. Some of that data is stored
or collected in a cloud environment in encrypted form, say data from a medical device or a smart home. But it shall
still be possible to derive important conclusions from it, for example to send immediate help to a patient suffering a
heart attack.
Innovative solutions are needed in this area of tension. The research in cryptography provides some highly
sophisticated tools for solving the like problems.
• Fully homomorphic encryption (FHE).
• Zero-Knowledge techniques, in particular: Non-interactive zero-knowledge proof (NIZKs).
• Secure multi-party computations (MPC).
• Anonymisation, TOR. Pseudonymization. Blinding.
• Weaker privacy notions, like differential privacy.

Prerequisites
Recommended:
Basic knowledge in cryptography (for example from MA-INF 1103) is highly recommended.
A profound mathematical background does help. In particular, precise mathematical formulation and reasoning are
important, but also topics like elementary number theory and discrete mathematics, especially lattices, are
interesting.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. Each student must present twice in the tutorial.
Master Computer Science — Universität Bonn 25

MA-INF 1224 Quantum Computing Algorithms

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Christian Bauckhage Prof. Dr. Christian Bauckhage

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Upon successful completion of this module, students should be able to describe fundamental concepts behind
working quantum algorithms.
Students acquire quantum computing programming know-how; based on knowledge and skills acquired, students
should be able to
• run quantum algorithms on (simulated) quantum computing platforms
• devise their own algorithms for optimization or classification problems that can be solved on quantum computers

Learning goals: soft skills

In the exercises, students can put their quantum computing knowledge into practice and realize small projects
involving the implementation of quantum algorithm. This requires teamwork; upon successful completion of the
module, students should be able to
• draft and implement basic quantum computing algorithms
• apply quantum computing (simulations) to test these algorithms
• prepare and give oral presentations about their work in front of an audience

Contents
quantum gate algorithms such as Deutsch-Jozsa, Bernstein-Vazirani, Simon, Shor, Grover; phase kick-back,
amplitude amplification; swap tests; Hamiltonian simulation, Trotterization, variational quantum computing for
optimization

Prerequisites
Required:
MA-INF 1107 - Foundations of Quantum Computing

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam
Ungraded coursework (required for admission to the exam)
The completion of regularly provided exercise sheets. The work can be done in groups of up to five students. A total
of 50% of the points must be achieved. Each student must present a solution to an exercise in the exercise sessions
once.
Forms of media
• lecture slides / lecture notes are made available online
• notebooks with programming examples are made available online
Master Computer Science — Universität Bonn 26

Literature
M.A. Nielsen, I.L Chuang, “Quantum Computation and Quantum Information”, Cambridge University Press, 10th
Anniversary edition, 2010
P. Wittek, “Quantum Machine Learning”, Academic Press, 2016
M. Schuld, F. Petruccione, “Machine Learning with Quantum Computers”, Springer, 2nd edition, 2021
S. Ganguly, “Quantum Machine Learning: An Applied Approach”, Apress, 2021
Master Computer Science — Universität Bonn 27

MA-INF 1225 Lab Exploring HPC technologies

Workload Credit points Duration Frequency

270 h 9 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Estela Suarez Prof. Dr. Estela Suarez
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Understanding a use case from complex code developed. Adapting and running applications to different kinds of
processing units, taking into account their specific architecture characteristic and programming environments.
Understanding and using parallel programming paradigms and high-level programming languages. Designing and
executing a benchmarking campaign. Using performance analysis tools, understanding performance bottlenecks and
measures to improve them. Software development skills and standards.

Learning goals: soft skills

Ability to analyze computational problems and to find creative and practical solutions; to examine one’s solutions
and results critically; to classify one’s own results into the state-of-the-art of the respective area; to produce good
quality software, prepare readable documentation of software and research results; to present, defend and discuss
design decisions and results in the team/group and to other students clearly and in accordance with academic
standards; to aim at long-range goals under limited resources; to work under pressure.

Contents
The students carry out a practical task (project) in High Performance Computing (HPC), including test of different
hardware architectures and software tools, documentation of the implemented software/system. Contents might be
adapted to a specific topic focus on different semesters. Example areas of work are: HPC systems: access/use of
compute resources; Use or simulation of different processor architectures; Software environment, performance
analysis tools; Parallel programming; Benchmarking tools/procedures; Performance of applications and scaling
behavior, strategies for improvement; Studying relation between architecture and application performance. Detailed
information on semester-specific focus topics will be given over the course website.

Prerequisites
Recommended:
• Knowledge of modern programming languages (C/C++, Python).
• Willingness to stay for at least 2 days per week during 4 weeks at the Jülich Supercomputing Centre, dates to be
discussed.
• Interest in High Performance Computing
• Recommended: Passed the exam of MA-INF 1108 Introduction to High Performance Computing: Architecture
Features and
Practical Parallel Programming (or its precursor MA-INF 1106).

Remarks
Registration first via direct mail communication with the lecturer, in order to identify level of previous knowledge
and suitable dates for the stay at JSC.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 2 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 28

Forms of media
Own laptop to connect and program on workstation or supercomputers.

Literature
• John L. Hennessy, David A. Patterson: Computer Architecture - A Quantitative Approach. Morgan Kaufmann
Publishers, 2012
• David A. Patterson, John L. Hennessy: Computer Organization and Design - The Hardware / Software Interface.
Morgan Kaufmann Publishers, 2013
• Message Passing Interface Forum: MPI: A Message-Passing Interface Standard, Version 3.1
• OpenMP Application Programming Interface, Version 4.5, November 2015
• Further semester-specific literature will be announced in the lab website.
Master Computer Science — Universität Bonn 29

MA-INF 1226 Applications of Computational Topology in Information

Theory

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Anne Driemel Dr. Felix Jonathan Boes, Dr. Benedikt Kolbe

Programme Mode Semester

M. Sc. Computer Science Optional 2-4.

Learning goals: technical skills

Upon successful completion of the module, students should have the ability to, in the treated topics, verify the
validity of propositions from original literature independently and to question research results critically. Students
acquire the competency to engage in independent study on current research topics.

Learning goals: soft skills

Social competence (communication, presenting one’s own solutions, goal-oriented discussions in teams), methodical
competence (analysis, abstraction, proofs), individual competence (commitment and willingness to learn, creativity,
endurance).

Contents
The course treats the usage of tools and algorithms from algebraic topology, particularly (co)homology theory, for
problems in computer science. Topics covered include: Symmetries as groups, (co)homology, equivariant cohomology,
topology in the modeling of the problem, algorithmic properties. The main focus is the connection of error correcting
codes, manifolds with involution and lattices.
Special features
The course will cover the use of advanced mathematical machinery (e.g. cohomology) from topology in applications
in computer science.

Prerequisites
Recommended:
• MA-INF1323 Computational Topology
• MA-INF1315 Lab Computational Geometry
• MA-INF1203 Discrete and Computational Geometry

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam
Ungraded coursework (required for admission to the exam)
Participation in an achievement test (midterm exam). At least 50% of the points much be achieved on this test.
Master Computer Science — Universität Bonn 30

MA-INF 1228 Graduate Seminar on Algorithms and Optimization

Workload Credit points Duration Frequency

180 h 6 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. László Végh Prof. Dr. László Végh

Programme Mode Semester

M. Sc. Computer Science Optional 2-4.

Learning goals: technical skills

Ability to undertake independent study of an advanced topic in discrete optimization using specialized literature.
Assessment, evaluation and presentation of results from algorithms and optimization. Didactic preparation and
presentation as a seminar talk and in the form of a manuscript covering the contents of the talk. Competence in
scientific discussions.
Learning goals: soft skills

Contents
A current, active research topic in algorithms and optimization chosen on a rotational basis will be treated in depth
by studying the relevant literature.

Prerequisites
Recommended:
• MA-INF 1102 - Combinatorial Optimization

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 4 60 T / 120 S 6

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 31

MA-INF 1229 Lab Advanced Algorithm Engineering

Workload Credit points Duration Frequency

270 h 9 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Petra Mutzel Prof. Dr. Petra Mutzel
Programme Mode Semester
M. Sc. Computer Science Optional 2-3.

Learning goals: technical skills

• Ability to develop high-quality algorithmic solutions by combining theoretical insights with practical engineering:
design and analyze algorithms, implement them efficiently, and evaluate them through sound experimental
methodology.
• Understanding and using performance engineering (benchmark design, profiling, optimization, and scalability
testing).
• Designing statistically sound experiments (controlled settings, fair and reproducible comparisons, meaningful
metrics) and interpreting results
critically.
• Produce scientifically clean documentation, including a structured written report and a reusable, well-documented
software artifact.
Learning goals: soft skills
• Collaborate with other students in small groups.
• Ability to scientifically present solutions and methods.
• Critical discussion of applied methods and techniques clearly and in accordance with academic standards.

Contents
The lab focuses on the full algorithm engineering cycle: from algorithmic design and analysis to efficient
implementation and reproducible experimental evaluation. Students work on selected combinatorial optimization
and algorithmic problems and develop exact and/or approximation algorithms and data structures. After a
structured literature review and definition of evaluation criteria, students design new algorithmic approaches or
adapt state-of-the-art methods. They then implement their solutions and build reproducible experimental pipelines
(benchmark selection, metric definition, parameter tuning, and controlled comparisons).
The lab concludes with a rigorous empirical study including profiling, optimization, scalability experiments, and a
scientifically documented report
and software artifact.
Prerequisites
Recommended:
Essential are knowledge of:
• Fundamental algorithms and algorithmic paradigms (e.g., graph algorithms, greedy algorithms, divide and
conquer, dynamic programming), data structures (e.g., balanced search trees, hash tables).
• Mathematical foundations of algorithm analysis (e.g., Big O notation, recurrence relations, proof techniques,
running-time analysis).
• Computational complexity (e.g., NP-hardness, reductions).
It is recommended to first complete the module
• MA-INF 1105 Algorithms for Data Analysis.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report
Master Computer Science — Universität Bonn 32

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
The relevant literature will be announced in time.
Master Computer Science — Universität Bonn 33

MA-INF 1230 Algorithmic Knot Theory

Workload Credit points Duration Frequency

180 h 6 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Anne Driemel Dr. Benedikt Kolbe, Dr. Felix Boes

Programme Mode Semester

M. Sc. Computer Science Optional 2-3.

Learning goals: technical skills

Upon successful completion of the module, students should have the ability to, in the treated topics, verify the
validity of propositions from original literature independently and to question research results critically. Students
acquire the competency to engage in independent study of current research topics.

Learning goals: soft skills

Social competence (communication, presenting one’s own solutions, goal-oriented discussions in teams), methodical
competence (analysis, abstraction, proofs), individual competence (commitment and willingness to learn, creativity,
endurance).

Contents
The course focuses on using geometric and topological methods to study knot theory in algorithmic computational
contexts and applications in both computer science and other natural sciences. The broached topics are studied
together with their mathematical foundations. The following is a tentative list of topics to be treated in the course.
Introduction to knot theory, the equivalence and unknot recognition problem, diagrammatic and surface
representations, the computation of knot polynomials, braid groups, enumerations of knots, applications of knot
theory, knots in 3-manifolds, 3-manifold recognition, the relevance of the Poincaré conjecture, Khovanov (and
Heegaard Floer) homology, and machine learning approaches.
Special features: The in-depth treatment of the topics requires substantial mathematical machinery, for which a high
degree of mathematical maturity is very helpful.

Prerequisites
Recommended:
• MA-INF 1323 - Computational Topology
• MA-INF 1315 - Lab Computational Geometry
• MA-INF 1203 - Discrete and Computational Geometry

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam
Ungraded coursework (required for admission to the exam)
Passing an ungraded midterm exam required for grade
Master Computer Science — Universität Bonn 34

MA-INF 1301 Algorithmic Game Theory

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Thomas Kesselheim Prof. Dr. Thomas Kesselheim, Senior Prof. Dr. Marek Karpinski

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Knowledge of fundamental results in (algorithmic) game theory and (algorithmic) mechanism design. Techniques
and methods related to mathematical modeling of strategic agents. Analyzing and designing systems of strategic
agents, with a focus on computational efficiency and performance guarantees.

Learning goals: soft skills

Presentation of solutions and methods, critical discussion of applied methods and techniques

Contents
• basic game theory
• computability and hardness of equilibria
• convergence of dynamics of selfish agents
• (bounds on the) loss of performance due to selfish behavior
• designing incentive-compatible auctions
• maximizing revenue
• designing mechanisms for stable and fair allocations without money

Prerequisites
Recommended:
Introductory knowledge of foundations of algorithms and complexity theory is essential.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam
Ungraded coursework (required for admission to the exam)
The completion of regularly provided exercise sheets. Each student must present a solution to an exercise in the
exercise sessions once.
Literature
• N. Nisan, T. Roughgarden, E. Tardos, V.V. Vazirani (ed.): Algorithmic Game Theory, Cambridge Univ. Press,
2007
• T. Roughgarden, Twenty Lectures on Algorithmic Game Theory, Cambridge Univ. Press, 2016
• A. Karlin, Y. Peres, Game Theory, Alive, AMS, 2017
• Y. Shoham, K. Leyton-Brown, Multiagent Systems, Cambridge Univ. Press, 2009
• D. M. Kreps: A Course in Microeconomic Theory, Princeton Univ. Press, 1990
• M. J. Osborne, A. Rubinstein: A Course in Game Theory, MIT Press, 2001
Master Computer Science — Universität Bonn 35

MA-INF 1307 Seminar Advanced Algorithms

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Thomas Kesselheim Prof. Dr. Anne Driemel, Prof. Dr. Thomas Kesselheim,
Prof. Dr. Heiko Röglin, PD Dr. Elmar Langetepe, Dr. Herman Haverkort

Programme Mode Semester

M. Sc. Computer Science Optional 3.

Learning goals: technical skills

Presentation of selected advanced topics in algorithm design and various applications

Learning goals: soft skills

Ability to perform individual literature search, critical reading, understanding, and clear didactic presentation

Contents
Advanced topics in algorithm design based on newest research literature

Prerequisites
none
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
The relevant literature will be announced in time.
Master Computer Science — Universität Bonn 36

MA-INF 1309 Lab Efficient Algorithms: Design, Analysis and

Implementation

Workload Credit points Duration Frequency

270 h 9 CP 1 semester at least every year

Module coordinator Lecturer(s)

Prof. Dr. Heiko Röglin Prof. Dr. Anne Driemel, Prof. Dr. Thomas Kesselheim,
Prof. Dr. Heiko Röglin, PD Dr. Elmar Langetepe, Dr. Herman Haverkort

Programme Mode Semester

M. Sc. Computer Science Optional 1-3.

Learning goals: technical skills

Ability to independently design, analyze and implement efficient algorithms and data structures for selected
computational problems

Learning goals: soft skills

Ability to analyze problems theoretically and to find creative and practical solutions; to examine one’s solutions and
results critically; to classify one’s own results into the state-of-the-art of the
respective area; to prepare readable documentation of software and research results; to present, defend and discuss
design decisions and results in the team/group and to other students clearly and in accordance with academic
standards; to collaborate constructively with others in small teams over a longer period of time; to aim at long-range
goals under limited resources; to work under pressure.

Contents
Design of efficient exact and approximate algorithms and data structures for selected computational problems.

Prerequisites
Recommended:
Knowledge of:
• fundamental algorithms and algorithmic paradigms (e.g., graph algorithms, greedy algorithms, divide and conquer,
dynamic programming), data structures (e.g., balanced search trees, hash tables)
• mathematical foundations of algorithm analysis (e.g., Big O notation, recurrence relations, proof techniques,
running-time analysis)
• computational complexity (e.g., NP-hardness, reductions)
It is recommended to take at least one of the following modules first:
• MA-INF 1102 Combinatorial Optimization
• MA-INF 1103 Cryptography
• MA-INF 1105 Algorithms for Data Analysis
• MA-INF 1107 Foundations of Quantum Computing
• MA-INF 1108 Introduction to High-Performance Computing: Architecture Features and Practical Parallel
Programming
• MA-INF 1201 Approximation Algorithms
• MA-INF 1202 Chip Design
• MA-INF 1203 Discrete and Computational Geometry
• MA-INF 1213 Randomized Algorithms and Probabilistic Analysis
• MA-INF 1218 Algorithms and Uncertainty
• MA-INF 1227 Hardness of Approximation
• MA-INF 1301 Algorithmic Game Theory
• MA-INF 1314 Online Motion Planning
• MA-INF 1323 Computational Topology

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9
Master Computer Science — Universität Bonn 37

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
The relevant literature will be announced in time.
Master Computer Science — Universität Bonn 38

MA-INF 1314 Online Motion Planning

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

PD Dr. Elmar Langetepe Prof. Dr. Rolf Klein, PD Dr. Elmar Langetepe

Programme Mode Semester

M. Sc. Computer Science Optional 1-4.

Learning goals: technical skills

To acquire fundamental knowledge on topics and methods in online motion planning

Learning goals: soft skills

Contents
Search and exploration in unknown environments (e.g., graphs, cellular environmwents, polygons, strets), online
algorithms, competitive analysis, competitive complexity,functional optimization, shortest watchman route, tethered
robots, marker algorithms, spiral search, approximation of optimal search paths.

Prerequisites
Recommended:
BA-INF 114 – Grundlagen der algorithmischen Geometrie

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam
Ungraded coursework (required for admission to the exam)
The completion of regularly provided exercise sheets. The work can be done in groups of up to three students. A
total of 25% of the points must be achieved. Each student must present a solution to an exercise in the exercise
sessions once.
Forms of media
Java applets of geometry lab

Literature
Scientific research articles will be recommended in the lecture.
Master Computer Science — Universität Bonn 39

MA-INF 1315 Lab Computational Geometry

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Anne Driemel Prof. Dr. Anne Driemel, PD Dr. Elmar Langetepe, Dr. Herman Haverkort

Programme Mode Semester

M. Sc. Computer Science Optional 2.

Learning goals: technical skills

Ability to design, analyze, implement and document efficient algorithms for selected problems in computational
geometry.

Learning goals: soft skills

Ability to properly present, defend and discuss design and implementation decisions, to document software according
to given rules and to collaborate with other students in small groups.

Contents
Various problems in computational geometry.

Prerequisites
none
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
The relevant literature will be announced in time.
Master Computer Science — Universität Bonn 40

MA-INF 1316 Lab Cryptography

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Dr. Michael Nüsken Dr. Michael Nüsken
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

The students will carry out a practical task (project) in the context of Cryptography, including test and
documentation of the implemented software/system.

Learning goals: soft skills

Ability to properly present and defend design decisions, to prepare readable documentation of software; skills in
constructively collaborating with others in small teams over a longer period of time; ability to classify ones own
results into the state-of-the-art of the resp. area.

Contents
Front of research topics in cryptography, in particular, related to fully homomorphic encryption, multi-party
computation, automated security verification.
The target of the lab is to understand how cryptography may work in one particular application that we are
choosing together. Ideally, we can come up with a novel solution for performing an unconsidered algorithm. We
study the tasks and tools, select algorithms, find a protocol, prototype an implemention, perform a security analysis,
present an evaluation.

Prerequisites
Recommended:
Good knowledge in cryptography is vital, eg. by
• MA-INF 1103 - Cryptography
• MA-INF 1223 - Privacy Enhancing Technologies
• MA-INF 1209 - Seminar Advanced Topics in Cryptography.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 41

MA-INF 1322 Seminar Focus Topics in High Performance Computing

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Estela Suarez Prof. Dr. Estela Suarez
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Enhanced and in-depth knowledge on topics and trends in the area of high performance computing.

Learning goals: soft skills

Acquire the competence to independently search for and study state-of-the-art scientific literature in depth, read
critically, identify the most relevant content, and assess research results in the context of the corresponding research
area; to execute peer-review processes, both to review work from others and to write rebuttal letters to reply
reviewer reports; to discuss research results with a knowledgeable scientific audience; to present prior work by others
in writing and in presentations with visual media in a way that adheres to academic standards, that is
well-structured and didactically effective, and that motivates the audience to participate; to formulate and accept
criticism; to manage one’s time with relatively open assignments and long-ranging deadlines.

Contents
General topics and trends in high performance computing, based on recent review and research literature

Prerequisites
Recommended:
MA-INF 1108 Introduction to High Performance Computing: Architecture Features and Practical Parallel
Programming

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
Literature and further information about this seminar will be announced in time in the website of lecturer.
Master Computer Science — Universität Bonn 42

MA-INF 1323 Computational Topology

Workload Credit points Duration Frequency

270 h 9 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Anne Driemel Prof. Dr. Anne Driemel, Dr. Benedikt Kolbe

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Knowledge of fundamental theorems and concepts in the area of computational topology in particular, persistent
homology and topological data analysis; design and analysis of combinatorial algorithms in topological contexts;
analysis of the complexity; to apply this knowledge autonomously to solving new problems and analysing new data
sets.
Learning goals: soft skills
Social competence (communication, presenting one’s own solutions, goal-oriented discussions in teams), methodical
competence (analysis, abstraction, proofs), individual competence (commitment and willingness to learn, creativity,
perseverance).

Contents
Fundamental concepts of relative homology and cohomology theory and persistence theory in computational
settings, category theory in this context, algorithms for the computation of (persistent) homology, (extended)
persistence modules and their decompositions, Morse theory, duality theorems, quiver representation theory, stability
of persistence diagrams and barcodes, algebraic stability, topological filtrations, multiparameter persistence,
invariants of persistence, topological data analysis, applications to shape pattern recognition, machine learning,
identification of geometric objects.

Prerequisites
none
Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam
Ungraded coursework (required for admission to the exam)
The completion of regularly provided exercise sheets. The work can be done in groups of up to two students. A total
of 50% of the points must be achieved. Each student must present a solution to an exercise in the exercise sessions
once.
Literature
• Herbert Edelsbrunner, John Harer (2010). Computational Topology: An Introduction. American Mathematical
Society.
• Steve Oudot (2015). Persistence Theory: From Quiver Representations to Data Analysis (Vol. 209). American
Mathematical Society.
• Magnus Bakke Botnan, Michael Lesnick (2022). An Introduction to Multiparameter Persistence.
• Allen Hatcher (2002). Algebraic Topology (Vol. 44). Cambridge University Press.
Master Computer Science — Universität Bonn 43

2 Graphics, Vision, Audio

MA-INF 2113 L2E2 6 CP Foundations of Audio Signal Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

MA-INF 2114 L2E4 6 CP Foundations of 4D/6D Object Capture for Virtual Environments . . . . . . . . . 45
MA-INF 2115 L4E2 9 CP Computational Medical Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47
MA-INF 2201 L4E2 9 CP Computer Vision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49
MA-INF 2202 L4E2 9 CP Computer Animation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50
MA-INF 2206 Sem2 4 CP Seminar Vision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
MA-INF 2207 Sem2 4 CP Seminar Graphics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
MA-INF 2208 Sem2 4 CP Seminar Audio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53
MA-INF 2209 L4E2 9 CP Advanced Topics in Computer Graphics I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
MA-INF 2210 Sem2 4 CP Seminar Computer Animation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 56
MA-INF 2212 L2E2 6 CP Pattern Matching and Machine Learning for Audio Signal Processing . . . . 57
MA-INF 2213 L3E1 6 CP Advanced Computer Vision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58
MA-INF 2214 L2E2 6 CP Computational Photography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
MA-INF 2215 Sem2 4 CP Seminar Digital Material Appearance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
MA-INF 2216 Lab4 9 CP Lab Visual Computing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
MA-INF 2219 Sem2 4 CP Seminar Visualization and Medical Image Analysis . . . . . . . . . . . . . . . . . . . . . . . 62
MA-INF 2220 Lab4 9 CP Lab Visualization and Medical Image Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . 63
MA-INF 2221 Sem2 4 CP Seminar Visual Computing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 64
MA-INF 2222 L4E2 9 CP Visual Data Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
MA-INF 2225 L2E2 6 CP Discrete Models for Visual Computing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66
MA-INF 2226 Lab4 9 CP Lab Geometry Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
MA-INF 2227 Lab4 9 CP Lab 3D Animation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 68
MA-INF 2228 Sem2 4 CP Seminar Vision and Graphics (Role-Based) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
MA-INF 2229 Sem2 4 CP Seminar Recent Advances in Geometry Processing . . . . . . . . . . . . . . . . . . . . . . . 70
MA-INF 2230 Lab4 9 CP Lab Computational Medical Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71
MA-INF 2307 Lab4 9 CP Lab Vision . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 73
MA-INF 2308 Lab4 9 CP Lab Graphics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74
MA-INF 2309 Lab4 9 CP Lab Audio . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 75
MA-INF 2310 L4E2 9 CP Advanced Topics in Computer Graphics II . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76
MA-INF 2311 Lab4 9 CP Lab Computer Animation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78
MA-INF 2312 L3E1 6 CP Image Acquisition and Analysis in Neuroscience . . . . . . . . . . . . . . . . . . . . . . . . . . 79
MA-INF 2316 Lab4 9 CP Lab Digital Material Appearance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80
MA-INF 2317 L2E2 6 CP Numerical Algorithms for Visual Computing and Machine Learning . . . . . . 81
MA-INF 2318 Sem2 4 CP Seminar Computational Medical Imaging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82
Master Computer Science — Universität Bonn 44

MA-INF 2113 Foundations of Audio Signal Processing

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

apl. Prof. Dr. Frank Kurth apl. Prof. Dr. Frank Kurth, Prof. Dr. Michael Clausen

Programme Mode Semester

M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

• Introduction to basic concepts of analog and digital signal processing: Acquire basic knowledge on modeling and
representing audio content; learn fundamental concepts of analog and digital signal processing, in particular
mathematical models of signal spaces and apply them to the audio domain; learn methods for analog to digital
conversion, frequency analysis, time-frequency analysis and digital filtering.
• Applications in the field of Audio Signal Processing: Learn typical application domains of audio signal processing
techniques and how to apply the acquired methods in solving applications problems from those domains. Important
examples are basic signal manipulation and filtering.
• Solving basic Signal Processing Problems: Learn basic signal processing algorithms for performing the Fourier
Transform and a time-frequency analysis, as well as for performing filter operations and fundamental types of signal
manipulations.
• Implementing basic Signal Processing Algorithms using state-of-the-art software frameworks: In the exercises, the
introduced methods and algorithms have to be implemented and applied to basic applications problems. Hence
knowledge in the practical implementation of digital signal processing methods in standard programming
environments such as Python, Matlab or Octave is acquired.

Learning goals: soft skills

Capability to analyze; Time management; Presentation skills; Discussing own solutions and solutions of others, and
working in groups.

Contents
Theoretical introduction to analog and digital Signal Processing; Fourier Transforms; Analog to digital Conversion;
Digital Filters; Audio Signal Processing Applications; Filter banks; Windowed Fourier Transform; 2D-Signal
Processing

Prerequisites
Recommended:
Solid basic knowledge on Linear Algebra and Analysis on the level acquired in Bachelor in Computer Science
programmes, including proficiency in using complex numbers.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of two to four students. A
total of 50% of the points must be achieved.

Forms of media
Slides, Blackboard, Whiteboard
Master Computer Science — Universität Bonn 45

MA-INF 2114 Foundations of 4D/6D Object Capture for Virtual

Environments
Workload Credit points Duration Frequency
180 h 6 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Reinhard Klein Prof. Dr. Reinhard Klein, Dr. Patrick Stotko

Programme Mode Semester

M. Sc. Computer Science Optional 1-2.

Learning goals: technical skills

• Knowledge about 3D/4D/6D data capturing and how to apply state-of-the art models and scene representations to
effectively process these data
• Make proper use and integrate solutions in game engines like Unity and Unreal Engine and standard tools like
Blender for practical applications
• Development and realization of individual state-of-the-art graphics and vision approaches

Learning goals: soft skills

• Communicative Skills: Written and oral presentation of solutions, discussing ideas in small teams, and preparing
structured written documents.
• Self-Competences: include time management, goal-oriented work, the ability to analyze problems theoretically, and
finding practical solutions
• Social Skills: involves effective teamwork, collaborating with others, accepting and formulating criticism, and
critical examination of research results
• Practical Skills: ability to implement practical solutions, present and defend design decisions, and prepare readable
documentation of software or projects

Contents
This intensive course offers an overview of the latest techniques and trends in 3D/4D/6D visual data processing and
demonstrates how these basic concepts can be applied to game engines and standard graphics tools. The covered
topics will be:
• Foundations of Computer Graphics and Vision
• Use of Deep Learning techniques in visual data processing
• Data acquisition techniques for Graphics and Vision
• Human model representations
• Motion data processing
• Geometry processing techniques
• Differentiable rendering for 3D/4D/6D reconstruction and model optimization
• Neural Radiance Fields and Gaussian Splatting as efficient scene representations
• Dynamic scene representations

Prerequisites
none
Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 4 60 T / 45 S 3.5

Graded exams
Written exam in three parts

Ungraded coursework (required for admission to the exam)

Successful participation in the exercise requires a minimum of 50% correct unit tests for the programming
assignments in each 5-day period
Master Computer Science — Universität Bonn 46

Literature
Supplemental readings will be provided before the lecture starts.
Master Computer Science — Universität Bonn 47

MA-INF 2115 Computational Medical Imaging

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Thomas Schultz Prof. Dr. Shadi Albarqouni

Programme Mode Semester

M. Sc. Computer Science Optional 3.

Learning goals: technical skills

• Learn the fundamental paradigms and mathematical foundations of computational medical imaging.
• Understand image formation processes (e.g., CT/MRI physics), and how they translate into discrete reconstruction
algorithms.
• Master classical and modern algorithms for: Image Segmentation (e.g., region-growing, atlas-based, clustering;
CNN-based methods); Image Registration (e.g., affine, deformable, diffeomorphic; learning-based registration);
Feature Extraction & Quantitative Imaging (e.g., intensity- and texture-based radiomics); Machine-Learning &
Deep-Learning in Medical Imaging (e.g., UNet, GANs for medical tasks)
• Formulate imaging problems as optimization tasks; derive and implement relevant cost functions and solvers.
• Analyze performance measures and evaluation protocols used in research and clinical practice.

Learning goals: soft skills

• Develop social competences through group exercises and in-class discussions.
• Enhance communicative skills via in-lecture question sessions and short presentations.
• Write concise algorithmic descriptions in reports; present technical material to peers.

Contents
Lecture (4 h/week).
• Introduction to Medical Imaging Modalities (CT physics, MRI basics, ultrasound principles).
• Segmentation Techniques (graph-cuts, level sets, statistical shape models, deep segmentation).
• Registration & Alignment (rigid vs. nonrigid, mutual information, Demons, SyN, learning-based registration).
• Quantitative Imaging & Radiomics (feature engineering, dimensionality reduction, classifiers).
• Machine Learning Foundations (supervised vs. unsupervised, loss functions, cross-validation in medical data).
• Deep Learning Architectures (CNN fundamentals, U-Net, GANs, Vision Transformers applied to medical data).
• Uncertainty & Bias in Medical Imaging (Bayesian networks, domain adaptation, causal inference).
• Ethical, Regulatory & Clinical Translation (data privacy, FDA/EMA guidelines, explainability, fairness).
• Emerging Topics (federated learning in healthcare, multimodal imaging, large language models in radiology).
Exercises (2 h/week)
• Hands-on segmentation tasks with public datasets (e.g., Brain MRI Tumor Segmentation), compare classical vs.
CNN approaches.
• Registration exercise using SimpleITK (perform rigid and nonrigid registration, compute registration error).
• Implement a small end-to-end deep learning pipeline (data loading, model training, evaluation) for a simple
classification or segmentation task.
• Evaluation metrics: compute Dice coefficient, Hausdorff distance, ROC/AUC for a provided dataset.

Prerequisites
Recommended:
• Basic knowledge of linear algebra, analysis, Matlab, and Python.
• MA-INF 2312 – Image Acquisition and Analysis in Neuroscience (or equivalent knowledge of medical imaging
modalities) und/oder:
• MA-INF 2222 – Visual Data Analysis (or equivalent knowledge of statistical data analysis).

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5
Master Computer Science — Universität Bonn 48

Graded exams
Written exam
Ungraded coursework (required for admission to the exam)
Successful exercise participation
Master Computer Science — Universität Bonn 49

MA-INF 2201 Computer Vision

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Jürgen Gall Prof. Dr. Jürgen Gall

Programme Mode Semester

M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

Students will be able to understand and explain mathematical descriptions of methods in publications from
Computer Vision. Students will be able to implement the discussed Computer Vision algorithms, apply them, and
choose the right approach and hyper-parameters for a given problem.

Learning goals: soft skills

Productive work in small teams, development and realization of individual approaches and solutions, critical
reflection of competing methods, discussion in groups.

Contents
The class will cover a number of mathematical methods and their applications in computer vision. For example,
linear filters, edges, derivatives, Hough transform, segmentation, graph cuts, mean shift, active contours, level sets,
MRFs, expectation maximization, background subtraction, temporal filtering, active appearance models, shapes,
optical flow, 2d tracking, cameras, 2d/3d features, stereo, 3d reconstruction, 3d pose estimation, articulated pose
estimation, deformable meshes, RGBD vision.

Prerequisites
Recommended:
Basic knowledge of linear algebra, analysis, probability theory, Python programming

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to two students. A total
of 50% of the points must be achieved.

Literature
• R. Hartley, A. Zisserman: Multiple View Geometry in Computer Vision
• R. Szeliski: Computer Vision: Algorithms and Applications
• S. Prince: Computer Vision: Models, Learning, and Inference
Master Computer Science — Universität Bonn 50

MA-INF 2202 Computer Animation

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Björn Krüger Prof. Dr. Björn Krüger

Programme Mode Semester

M. Sc. Computer Science Optional 1-3.

Learning goals: technical skills

Students will learn fundamental paradigms used in computer animation. They will learn the mathematical
foundations and basic algorithms to solve problems in the areas of motion capturing, motion synthesis, and motion
analysis.

Learning goals: soft skills

Social competences (work in groups), communicative skills (written and oral presentation)

Contents
Fundamentals of computer animation; kinematics; representations of motions; motion capturing; motion editing;
motion synthesis; facial animations

Prerequisites
Recommended:
Basic knowldge of linear algebra, analysis, Matlab and Python

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam (30 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to three students. A
total of 50% of the points must be achieved.

Literature
• Dietmar Jackel, Stephan Neunreither, Friedrich Wagner: Methoden der Computeranimation, Springer 2006
• Rick Parent: Computer Animation: Algorithms and Techniques, Morgan Kaufman Publishers 2002
• Frederic I. Parke , Keith Waters: Computer Facial Animation. A K Peters, Ltd. 199
• Grünvogel Stefan, Einführung in die Computer Animation, Springer 2024
Master Computer Science — Universität Bonn 51

MA-INF 2206 Seminar Vision

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Jürgen Gall Prof. Dr. Jürgen Gall

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Ability to understand new research results presented in original scientific papers.

Learning goals: soft skills

Ability to present and to critically discuss these results in the framework of the corresponding area.

Contents
Current conference and journal papers.

Prerequisites
Recommended:
MA-INF 2201 – Computer Vision or MA-INF 2213 - Advanced Computer Vision

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 52

MA-INF 2207 Seminar Graphics

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Reinhard Klein Prof. Dr. Reinhard Klein
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Ability to understand new research results presented in original scientific papers.

Learning goals: soft skills

Ability to present and to critically discuss these results in the framework of the corresponding area.

Contents
Current conference and journal papers.

Prerequisites
Recommended:
Mathematical background (multidimensional analysis and linear algebra, basic numerical methods)
Basic knowledge in Computer Graphics

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 53

MA-INF 2208 Seminar Audio

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

apl. Prof. Dr. Frank Kurth apl. Prof. Dr. Frank Kurth, Dr. Michael Clausen

Programme Mode Semester

M. Sc. Computer Science Optional 2.

Learning goals: technical skills

Ability to understand new research results presented in original scientific papers.

Learning goals: soft skills

Ability to present and to critically discuss these results in the framework of the corresponding area.

Contents
Current conference and journal papers in the area of audio signal processing.

Prerequisites
Recommended:
MA-INF 2113 - Audio Signal Processing

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 54

MA-INF 2209 Advanced Topics in Computer Graphics I

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Reinhard Klein Prof. Dr. Reinhard Klein
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Analytical formulation of problems related to rendering. Knowledge of principles, techniques and algorithms to
• recognize and understand the physical quantities of light transport
• explain a range of surface and volumetric material models
• explain the rendering and radiative transfer equations
• design and implement methods to solve these equations, especially Monte Carlo methods
• Assess / Evaluate the performance and conceptual limits of the implemented simulation code

Learning goals: soft skills

Based on the knowledge and skills acquired students should be able to
• read and judge current scientific literature in the area of rendering
• identify the major literature concerning a given problem in rendering and gain an overview of the current state of
the art
• discuss problems concerning rendering with researchers from different application fields
• present, propose and communicate different solutions and work in a team to solve a rendering problem

Contents
This course introduces the basic physical quantities as well as the mathematical and algorithmic tools required to
understand and simulate the light interaction with objects and different materials in a 3D scene. We will discuss how
to solve the mathematical problem numerically in order to create realistic images. Advanced topics include
participating media, material models for sub-surface light transport, and Markov Chain Monte Carlo Methods.
Topics among others will be
• rendering and radiative transfer equation
• methods and algorithms to solve these equations, radiosity, Monte Carlo, photon mapping
• analytical and data driven surface and subsurface material models, especially BRDF, BSSRDF models
• differentiable rendering
In addition, results from state-of-the-art research will be presented.

Prerequisites
Recommended:
Recommended but not enforced: basic knowledge in computer graphics, (numerical) analysis and linear algebra,
C++

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to two students. A total
of 50% of the points must be achieved. For 70% of the exercise sheets, 50% of the points must be achieved for each
sheet. The exercises are divided into theoretical and practical exercises, and the points to be achieved apply
separately to both categories.
Master Computer Science — Universität Bonn 55

Literature
• M. Pharr, W. Jakob, and G. Humphreys, Physically Based Rendering: From Theory to Implementation (3rd
edition), 2018
• L. Szirmay-Kalos: Monte-Carlo Methods in Global Illumination, Institute of Computer Graphics, Vienna
University of Technology, Vienna, 1999 URL: [Link]
• P. Dutre, K. Bala, P. Bekaert: Advanced Global Illumination, 2nd ed., B&T, 2006
• D’Eon, Eugene. A Hitchhiker’s Guide to Multiple Scattering, 2016
Master Computer Science — Universität Bonn 56

MA-INF 2210 Seminar Computer Animation

Workload Credit points Duration Frequency

120 h 4 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Reinhard Klein Prof. Dr. Björn Krüger

Programme Mode Semester

M. Sc. Computer Science Optional 2.

Learning goals: technical skills

Ability to understand new research results presented in original scientific papers.

Learning goals: soft skills

Ability to present and to critically discuss these results in the framework of the corresponding area.

Contents
Current conference and journal papers.

Prerequisites
Recommended:
At least 1 of the following:
MA-INF 2202 – Computer Animation
MA-INF 2311 – Lab Computer Animation

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 57

MA-INF 2212 Pattern Matching and Machine Learning for Audio Signal
Processing

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

apl. Prof. Dr. Frank Kurth apl. Prof. Dr. Frank Kurth

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

• Introduction into selected topics of digital signal processing: Acquire basic knowledge on representing and
manipulating 1D-time series. Learn basic methods for time-frequency analysis and signal processing methods for
feature extraction.
• Applications in the field of Audio Signal Processing: Learn typical application domains of audio signal processing
techniques and how to apply the acquired methods in solving applications problems. Important examples are
filtering, signal/object detection and classification tasks.
• Methods of Automatic Pattern Recognition and Machine Learning: Learn methods for Feature Extraction,
Automatic Pattern Recognition and Machine Learning for the context of Audio Signal Processing. Be able to apply
those fundamental methods (method list: see "Contents" section) in particular for solving applications tasks.

Learning goals: soft skills

Audio Signal Processing Applications; Extended programming skills for signal processing applications; Capability to
analyze; Time management; Presentation skills; Discussing own solutions and solutions of others, and working in
groups.

Contents
The lecture is presented in modular form, where each module is motivated from the application side. The presented
topics are: Windowed Fourier transforms; Audio Identification; Audio Matching; Signal Classification; Applications
of ML/DL concepts, in particular HMMs, SVMs and Neural Networks to Audio Signal processing tasks.

Prerequisites
Recommended:
Solid basic knowledge on Linear Algebra, Analysis and Stochastics, including proficiency in using complex numbers.
Having attended MA-INF 2113 Foundations of Audio Signal Processing is highly recommended, as fundamental
material from (Digital) Signal Processing and Audio Processing are introduced there in depth.
Basic knowledge in time series data analysis is helpful but not mandatory.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of two to four students. A
total of 50% of the points must be achieved.

Forms of media
Slides, Blackboard, Whiteboard
Master Computer Science — Universität Bonn 58

MA-INF 2213 Advanced Computer Vision

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Jürgen Gall Prof. Dr. Jürgen Gall

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Students will be able to implement the discussed machine learning algorithms for Computer Vision, apply them, and
choose the right approach and hyper-parameters for a given problem.

Learning goals: soft skills

Productive work in small teams, development and realization of individual approaches and solutions, critical
reflection of competing methods, discussion in groups.

Contents
The class will cover a number of computer vision applications. For example, image classification, object detection,
action recognition, pose estimation, face analysis, tracking, image synthesis, vision-language models.

Prerequisites
Recommended:
MA-INF 2201 – Computer Vision

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 3 45 T / 45 S 3 S = independent study
Exercises 1 15 T / 75 S 3

Graded exams
Oral exam (20 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to two students. A total
of 50% of the points must be achieved.
Master Computer Science — Universität Bonn 59

MA-INF 2214 Computational Photography

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Matthias Hullin Prof. Dr. Matthias Hullin
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Foundations in optics and image sensors. Signal processing and inverse problems in imaging. Color spaces and
perception. Image alignment and blending. High-dimensional representations of light transport (light fields,
reflectance fields, reflectance distributions). Computational illumination.

Learning goals: soft skills

• to read and understand current literature in the field
• to implement standard computational photography techniques
• to propose and implement solutions to a given problem
• to follow good scientific practice by planning, documenting and communicating their work

Contents
• Image sensors
• Optics
• Panoramas
• Light fields
• Signal processing and inverse problems
• Color, perception and HDR
• Reflectance fields and light transport matrices

Prerequisites
Required:
Basic knowledge in computer graphics, data structures, multidimensional analysis und linear algebra, numerical
analysis and numerical linear algebra, C++ or MATLAB

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam
Ungraded coursework (required for admission to the exam)
(i) The completion of regularly provided exercise sheets. The work can be done in groups of up to two students. A
total of 50% of the points must be achieved. The exercises are divided into theoretical and practical exercises, and
the points to be achieved apply separately to both categories. Each student must present a solution to an exercise in
the exercise sessions twice. (ii) The completion of a programming project. The work is done in groups of two to four
students, depending on the total number of students taking the course. The results of the programming project must
be presented in class.
Master Computer Science — Universität Bonn 60

MA-INF 2215 Seminar Digital Material Appearance

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Matthias Hullin Prof. Dr. Matthias Hullin
Programme Mode Semester
M. Sc. Computer Science Optional 2.

Learning goals: technical skills

Ability to understand new research results presented in original scientific papers.

Learning goals: soft skills

Ability to present and to critically discuss these results in the framework of the corresponding area.

Contents
Current conference and journal papers

Prerequisites
none
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 61

MA-INF 2216 Lab Visual Computing

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Florian Bernard Prof. Dr. Florian Bernard
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

• Getting into a selected topic of visual computing
• Implementation and practical application of current visual computing methods
• Experimental evaluation and visualisation of results
• Scientific research and writing

Learning goals: soft skills

• self-organisation
• ability to analyze problems theoretically and to find creative and practical solutions
• critical thinking: examine one’s solutions and results critically
• to classify own results into the state-of-the-art of the respective area
• to prepare readable documentation of software and research results
• to present, defend and discuss design decisions and results in the team/group and to other students clearly and in
accordance with academic standards
Contents
This lab introduces visual computing methods and applications. You will get a chance to study the methods in
depth by implementing them and running experiments. At the end of the semester, you will present the method,
give a short demonstration and hand in a report describing the method and experimental outcomes. Potential topics
include deep learning (e.g. graph neural networks, unsupervised learning, 3D deep learning), mathematical
optimization (e.g. linear/convex/non-convex programming, graph-based algorithms) and other methods involving
mathematical modeling of visual computing problems.

Prerequisites
Recommended:
Basic knowledge in mathematics (e.g. linear algebra, calculus, optimization) and programming (e.g. python, in
particular pytorch or tensorflow, C++, or Matlab). In addition:
• MA-INF 2317: Numerical Algorithms for Visual Computing and Machine Learning, or
• MA-INF 2225: Discrete Models for Visual Computing

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 62

MA-INF 2219 Seminar Visualization and Medical Image Analysis

Workload Credit points Duration Frequency

120 h 4 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Thomas Schultz Prof. Dr. Thomas Schultz
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Enhanced and in-depth knowledge in specialized topics in the area of visualization and medical image analysis.

Learning goals: soft skills

Acquire the competence to independently search for and study state-of-the-art scientific literature in depth, read
critically, identify the most relevant content, and assess research results in the context of the corresponding research
area; to discuss research results with a knowledgeable scientific audience; to present prior work by others in writing
and in presentations with visual media in a way that adheres to academic standards, that is well-structured and
didactically effective, and that motivates the audience to participate; to formulate and accept criticism; to manage
one’s time with relatively open assignments and long-ranging deadlines.

Contents
Recent research topics in visualization and medical image analysis based on journal and conference publications.
Relevant journals include Medical Image Analysis, IEEE Transactions on Medical Imaging, IEEE Transactions on
Visualization and Computer Graphics; relevant conferences include Medical Image Computing and
Computer-Assisted Intervention (MICCAI), IEEE/CVF Computer Vision and Pattern Recognition (CVPR) IEEE
VIS, EuroVis.

Prerequisites
Recommended:
At least one of the following:
• MA-INF 2222 – Visual Data Analysis
• MA-INF 2312 – Image Acquisition and Analysis in Neuroscience

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 63

MA-INF 2220 Lab Visualization and Medical Image Analysis

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Thomas Schultz Prof. Dr. Thomas Schultz
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Students acquire a deep understanding of a specific problem in visualization and medical image analysis, and
technical knowledge about state-of-the-art algorithmic approaches to solving it. This involves problem identification;
data processing; selection, design, implementation, and application of suitable algorithms; communication of results.

Learning goals: soft skills

Ability to analyze problems theoretically and to find creative and practical solutions; to examine one’s solutions and
results critically; to classify one’s own results into the state-of-the-art of the
respective area; to prepare readable documentation of software and research results; to present, defend and discuss
design decisions and results in the team/group and to other students clearly and in accordance with academic
standards; to collaborate constructively with others in small teams over a longer period of time; to aim at long-range
goals under limited resources.

Contents
The students will carry out a practical task (project) in the context of data visualization and visual analytics or
medical image analysis, including test and documentation of the implemented software/system. Projects are often
based on journal and conference publications. Relevant journals include Medical Image Analysis, IEEE Transactions
on Medical Imaging, IEEE Transactions on Visualization and Computer Graphics; relevant conferences include
Medical Image Computing and Computer-Assisted Intervention (MICCAI), IEEE/CVF Computer Vision and
Pattern Recognition (CVPR) IEEE VIS, EuroVis.

Prerequisites
Recommended:
At least one of the following:
• MA-INF 2222 – Visual Data Analysis
• MA-INF 2312 – Image Acquisition and Analysis in Neuroscience.
A solid background in programming is required, preferably in Python. Most projects also require basic knowledge in
linear algebra, calculus, probability theory, and/or numerical algorithms.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 64

MA-INF 2221 Seminar Visual Computing

Workload Credit points Duration Frequency

120 h 4 CP 1 semester at least every year

Module coordinator Lecturer(s)

Prof. Dr. Florian Bernard Prof. Dr. Florian Bernard
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Ability to understand new research results presented in original scientific papers.

Learning goals: soft skills

Ability to present and to critically discuss these results in the framework of the corresponding area.

Contents
Current conference and journal papers.

Prerequisites
Required:
No formal requirements. Participants are expected to have some previous exposure to at least one of the following:
- visual computing (e.g. computer vision, computer graphics, 3D shape analysis, image analysis, etc.),
- mathematical optimisation (e.g. combinatorial/continuous, convex/non-convex, etc.), or
- machine learning.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 65

MA-INF 2222 Visual Data Analysis

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Thomas Schultz Prof. Dr. Thomas Schultz
Programme Mode Semester
M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

Ability to design, implement, and make proper use of systems for visual data analysis. Knowledge of algorithms and
techniques for the visualization of multi-dimensional data, graphs, as well as scalar, vector, and tensor fields.

Learning goals: soft skills

Productive work in small teams, self-dependent solution of practical problems in the area of visual data analysis,
critical reflection on visualization design, presentation of solution strategies and implementations, self management

Contents
This class provides a broad overview of principles and algorithms for data analysis via interactive visualization.
Specific topics include perceptual principles, color spaces, visualization analysis and design, integration of visual with
statistical data analysis and machine learning, as well as specific algorithms and techniques for the display of
multidimensional data, dimensionality reduction, graphs, geospatial data, neural networks, as well as scalar, vector
and tensor fields.
Prerequisites
Recommended:
Students are recommended to have a basic knowledge in linear algebra and calculus, as well as proficiency in
programming.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to three students. A
total of 50% of the points must be achieved. Each student must present a solution to an exercise in the exercise
sessions once.
Literature
• A.C. Telea, Data Visualization: Principles and Practice. CRC Press, Second Edition, 2015
• M. Ward et al., Interactive Data Visualization: Foundations, Techniques, and Applications. CRC Press, 2010
• T. Munzner, Visualization Analysis and Design, A K Peters, 2015
Master Computer Science — Universität Bonn 66

MA-INF 2225 Discrete Models for Visual Computing

Workload Credit points Duration Frequency

180 h 6 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Florian Bernard Prof. Dr. Florian Bernard
Programme Mode Semester
M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

• Ability to implement basic visual computing algorithms, understanding their strengths and shortcomings
• Mathematical modelling of computational problems in visual computing
• Gain an intuition which algorithm is best applied for which problem in visual computing, so that practical
problems in these areas can be solved

Learning goals: soft skills

• Problem solving skills: ability to identify and utilise analogies between new problems and previously seen ones
• Analytical and abstract thinking: develop a general intuition of computational problems, being able to adopt
different perspectives of particular concepts

Contents
This module focuses on discrete models that frequently occur in the field of visual computing (VC). In addition to
algorithms, this module will also cover modelling aspects that are relevant for solving practical problems in VC. The
contents include:
• Graph-based models (e.g. linear and quadratic assignment, network flows, product graph formalisms, dynamic
programming and their application)
• Continuous algorithms for discrete problems (e.g. convex & spectral relaxations, projection methods,
path-following and their application)
• Deep Learning for discrete domains (e.g. differentiable programming, graph neural networks, deep learning on
meshes)

Prerequisites
Recommended:
Participants are expected to have a high level of mathematical maturity (in particular, a good working knowledge of
linear algebra and calculus/analysis is essential). A basic understanding of graph theory is advantageous.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam
Ungraded coursework (required for admission to the exam)
The completion of regularly provided exercise sheets. The work can be done in groups of up to two students. A total
of 50% of the points must be achieved.
Master Computer Science — Universität Bonn 67

MA-INF 2226 Lab Geometry Processing

Workload Credit points Duration Frequency

270 h 9 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Jun. Prof. Dr. Zorah Lähner Prof. Dr. Zorah Lähner
Programme Mode Semester
M. Sc. Computer Science Optional 1-3.

Learning goals: technical skills

Ability to handle complex geometric data types; to extract implementation details from research publications; to
implement and visualize geometric data.

Learning goals: soft skills

Ability to analyze problems theoretically and to find creative and practical solutions; to examine one’s solutions and
results critically; to classify one’s own results into the state-of-the-art of the respective area; to prepare readable
documentation of software and research results; to present, defend and discuss design decisions and results in the
team/group and to other students clearly and in accordance with academic standards; to collaborate constructively
with others in small teams over a longer period of time.

Contents
Mesh deformation, point cloud meshing, pytorch3D, shape correspondence, reconstruction, 2D-to-3D transfer. This
lab introduces methods and applications in the field of geometry processing. You will get a chance to study the
methods in depth by implementing them and running experiments. At the end of the semester, you will present the
method, give a short demonstration and hand in a report describing the method and experimental outcomes.

Prerequisites
none
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 68

MA-INF 2227 Lab 3D Animation

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Ina Prinz Prof. Dr. Ina Prinz
Programme Mode Semester
M. Sc. Computer Science Optional 1-3.

Learning goals: technical skills

The students will learn to carry out a practical task (project) in the context of 3D animation, containing modelling,
preparing a screenplay, realizing an animation related to real physical laws, rendering and creating a video.

Learning goals: soft skills

Ability to properly present and defend design decisions, to prepare readable documentation of software; skills in
constructively collaborating with others in small teams over a longer period of time; ability to classify ones own
results into the state-of-the-art of the resp. area.

Contents
Varying selected topics close to current research in the area of the history of computing and the mechanization of
computing as well as deep understanding of mechanical and technical functions and its presentation in a
representative 3D animation video, contains technical visualization and didactic skills.

Prerequisites
Recommended:
For students who did not take BA-INF 108 Geschichte des maschinellen Rechnens I and BA-INF 126 Geschichte des
maschinellen Rechnens II in their Bachelor’s studies, recommended reading includes:
• Aspray, W.: Computing before Computers. Ames, 1990.
• Bauer, Friedrich L.: Origins and Foundations of Computing. Berlin 2010.
• Ceruzzi, Paul E.: A History of Modern Computing. Cambridge, 2003.
• Goldstine, H.: The Computer from Pascal to von Neumann. Princeton, 1972.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report, presentation of the video

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 69

MA-INF 2228 Seminar Vision and Graphics (Role-Based)

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

Jun. Prof. Dr. Zorah Lähner Prof. Dr. Zorah Lähner
Programme Mode Semester
M. Sc. Computer Science Optional 2-4.

Learning goals: technical skills

• reading and understanding of research publications in the area of computer vision and computer graphics
• learning about different roles in the research community and taking their point of view

Learning goals: soft skills

• Critical thinking: ability to put research into wider context and analyze it from different perspectives
• Communication: oral and written presentation of scientific content, high level discussion about a new topic
• Self-Competence: time management, focusing on essential aspects , creativity

Contents
Students will study a variety of publications in the area of computer vision and graphics, and will be assigned a
specific role which determines how to interact with the work.
The roles include but are not limited to:
• Scientific Peer Reviewer
• Academic Researcher
• Archaeologist (putting the paper into context regarding previous and subsequent work)
• Industry Practitioner

Prerequisites
Recommended:
A background in visual computing through lectures from the Graphics, Vision, Audio subfield is highly
recommended.
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6). Participation in each type of
assigned role in-presence at least once.
Master Computer Science — Universität Bonn 70

MA-INF 2229 Seminar Recent Advances in Geometry Processing

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

Jun. Prof. Dr. Zorah Lähner Prof. Dr. Zorah Lähner
Programme Mode Semester
M. Sc. Computer Science Optional 2-4.

Learning goals: technical skills

This module examines recent topics in geometry processing. The students learn to do independent, in-depth study of
state-of-the-art scientific literature, discuss them with their peers and present in a form suited for a scientific
audience.
Learning goals: soft skills
Communication skills: oral and written presentation of scientific content
Self-competence: the ability to analyze problems, time management, creativity

Contents
Algorithmic and learning-based methods for geometry processing, including typical applications like shape
correspondence, 3D reconstruction, geometry evaluation, differential geometry, statistical modeling as well differences
for methods using implicit and explicit geometry representations.

Prerequisites
Recommended:
MA-INF 2310 Advanced Topics in Computer Graphics II

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 71

MA-INF 2230 Lab Computational Medical Imaging

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Thomas Schultz Prof. Dr. Shadi Albarqouni

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

• Acquire a deep understanding of a selected problem in computational medical imaging: from problem identification
to data processing, algorithm design, implementation, and evaluation.
• Implement, test, and validate state-of-the-art image-analysis algorithms (e.g., segmentation, registration,
reconstruction, classification) on real medical data.
• Develop proficiency in using libraries and toolkits commonly used in medical imaging (e.g., ITK, SimpleITK,
PyTorch for deep learning, nibabel/dipy).
• Learn to document software systematically (API documentation, code comments, unit tests) and to generate
reproducible experiments (version control, containerization).

Learning goals: soft skills

• Collaborate effectively in small groups over the entire semester, dividing tasks (e.g., data curation, algorithmic
implementation, evaluation).
• Prepare clear, concise technical documentation and design rationales.
• Present project progress and final results in oral form (midterm demo, final demo).
• Critically assess one’s own results and position them within the state-of-the-art in computational medical imaging.

Contents
Project-based lab: Students work in groups (max 3 per group) to carry out a semester-long project in computational
medical imaging. Possible project topics (chosen or proposed by the instructors) include:
• Anatomical Structure Segmentation (e.g., multi-atlas, CNN-based)
• Deformable Registration (e.g., classic demons, learning-based registration)
• Lesion/Anomaly Detection (e.g., out-of-distribution detection in X-ray)
• Radiomics & Quantitative Biomarker Extraction (e.g., texture features, deep radiomics)
• Uncertainty Estimation in Deep Learning (e.g., MC-Dropout, Bayesian networks)
• Causality & Bias Analysis in Medical Datasets (e.g., domain shift, fairness)
Each group will:
• Identify a well-defined problem and collect or be provided a suitable dataset (publicly available or hospital data
under ethics approval).
• Perform a literature review to select baseline methods.
• Implement baseline and at least one novel or improved approach.
• Evaluate quantitatively (e.g., Dice, Hausdorff, MSE, classification accuracy) and qualitatively (expert review where
possible)
• Document the software (README, API docs) and provide a reproducible environment (Docker/Singularity).
• Present midterm progress (oral demonstration + short written summary) and final results (oral + poster or full
report).

Prerequisites
Recommended:
At least one of the following:
• MA-INF 2222 – Visual Data Analysis.
• MA-INF 2312 – Image Acquisition and Analysis in Neuroscience.
• MA-INF 2220 – Lab Visualization and Medical Image Analysis (or equivalent).
A solid background in programming (Python and/or C++), basic linear algebra, calculus, probability theory, and/or
numerical algorithms is required. Familiarity with deep learning frameworks is highly recommended.
Master Computer Science — Universität Bonn 72

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation (final demo + Q&A, 20 minutes per group). Written report (12–16 pages, including background,
methods, results, discussion).

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 73

MA-INF 2307 Lab Vision

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Jürgen Gall Prof. Dr. Jürgen Gall

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

The students will carry out a practical computer vision task (project).

Learning goals: soft skills

Ability to properly present and defend design decisions, to prepare readable documentation of software; skills in
constructively collaborating with others in small teams over a longer period of time; ability to classify ones own
results into the state-of-the-art of the resp. area

Contents
Computer Vision: research topics and applications

Prerequisites
Required:
Good C++ or Python programming skills
Recommended:
MA-INF 2201 – Computer Vision or MA-INF 2213 - Advanced Computer Vision

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 74

MA-INF 2308 Lab Graphics

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Reinhard Klein Prof. Dr. Reinhard Klein
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

The students will learn to carry out a practical task (project) in the context of geometry processing, rendering,
scientific visualization or human computer interaction, including test and documentation of the implemented
software/system.

Learning goals: soft skills

Ability to properly present and defend design decisions, to prepare readable documentation of software; skills in
constructively collaborating with others in small teams over a longer period of time; ability to classify ones own
results into the state-of-the-art of the resp. area

Contents
Varying selected topics close to current research in the area of geometry processing, rendering, scientific visualization
or human computer interaction.

Prerequisites
Recommended:
At least one of the following:
• MA-INF 1108 Introduction to High Performance Computing
• MA-INF 2202 Computer Animation
• MA-INF 2222 Visual Data Analysis

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 75

MA-INF 2309 Lab Audio

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

apl. Prof. Dr. Frank Kurth apl. Prof. Dr. Frank Kurth, Prof. Dr. Michael Clausen

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Proficiency in implementing signal processing concepts introduced in selected scientific publications or research
reports. Proficiency in collecting and maintaining data sets, in particular signals and corresponding metadata, and
performing scientific evaluations of signal processing methods based on data sets and implemented algorithms.

Learning goals: soft skills

Ability to analyze problems theoretically and to find creative and practical solutions; to examine one’s solutions and
results critically; to classify one’s own results into the state-of-the-art of the respective area; to prepare readable
documentation of software and research results; to present, defend and discuss design decisions and results in the
team/group and to other students clearly and in accordance with academic standards; to collaborate constructively
with others in small teams over a longer period of time; to pursue long-range goals under a given resource budget.

Contents
In the lab a medium-sized programming project related to digital audio signal processing has to be solved during the
period of one semester. For this, initial literature, usually in form of one or two scientific papers or reports, will be
provided at the beginning of the lab. Also, resources regarding the audio signal data to be used, are given. Typical
programming tasks consist of implementing either general signal processing algorithms such as fundamental
frequency estimation or of implementing algorithms for solving application problems such as speaker detection or
classification. For participants with interest in topics of pattern recognition, machine and deep learning,
programming projects from those areas, with application to audio processing, can be selected.

Prerequisites
Recommended:
Solid basic proficiency in imperative programming (e.g. knowledge of C/C++, Java, Python). Knowledge of the
material from MA-INF 2113 Foundations of Audio Signal Processing is highly recommended. Knowledge of material
from MA-INF 2212 Pattern Matching and Machine Learning for Audio is helpful but not necessary.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 76

MA-INF 2310 Advanced Topics in Computer Graphics II

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Reinhard Klein Prof. Dr. Reinhard Klein
Programme Mode Semester
M. Sc. Computer Science Optional 3.

Learning goals: technical skills

Analytical formulation of problems related to geometry processing:
• apply methods of geometry processing
• apply basic concepts of statistical shape analysis and shape spaces to real world applications
• Design and implement novel application software in this area

Learning goals: soft skills

Based on the knowledge and skills acquired students should be able to
• read and judge current scientific literature in the area of geometry processing and gain an overview of the current
state of the art
• identify the major literature relevant for solving a given problem in geometry processing
• present, propose and communicate different solutions and work in a team to solve geometry processing problems
• discuss geometry processing problems with researchers from different application fields

Contents
This course will first introduce the mathematical and algorithmic tools required to represent, model, and process 3D
geometric objects. The second part discusses the latest mathematical, algorithmic, and statistical tools required for
the analysis and modeling of 3D shape variability, which can facilitate the creation of 3D models. Topics among
others will be
• classical and discrete differential geometry of curves and surfaces
• mesh data structures and generation of meshes from point clouds
• Laplacian operator and optimization techniques with applications to denoising, smoothing, decimation, shape
fitting, shape descriptors, geodesic distances
• parameterization and editing of surfaces
• point cloud registration
• correspondences
• shape spaces and statistical shape analysis
In addition, results from state-of-the-art research will be presented.

Prerequisites
none
Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam
Ungraded coursework (required for admission to the exam)
The completion of regularly provided exercise sheets. The work can be done in groups of up to two students. A total
of 50% of the points must be achieved. For 70% of the exercise sheets, 50% of the points must be achieved for each
sheet. The exercises are divided into theoretical and practical exercises, and the points to be achieved apply
separately to both categories.
Master Computer Science — Universität Bonn 77

Literature
• M. Botsch, L. Kobbelt, M. Pauly, P. Alliez, B. Levy, Polygon Mesh, Processing, A K Peters, 2010
• Laga, Hamid, Yulan Guo, Hedi Tabia, Robert B. Fisher, and Mohammed Bennamoun. 3D Shape analysis:
fundamentals, theory, and applications. John Wiley & Sons, 2018.
• Solomon, Justin. Numerical Algorithms. Textbook published by AK Peters/CRC Press, 2015
Master Computer Science — Universität Bonn 78

MA-INF 2311 Lab Computer Animation

Workload Credit points Duration Frequency

270 h 9 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Reinhard Klein Prof. Dr. Björn Krüger

Programme Mode Semester

M. Sc. Computer Science Optional 3.

Learning goals: technical skills

The students will carry out a practical task (project) in the context of computer animation, including test and
documentation of the implemented software/system.

Learning goals: soft skills

Ability to properly present and defend design decisions, to prepare readable documentation of software; skills in
constructively collaborating with others in small teams over a longer period of time; ability to classify ones own
results into the state-of-the-art of the resp. area

Contents
Varying selected topics close to current research in the area of computer animation.

Prerequisites
Recommended:
At least 1 of the following:
MA-INF 2202 – Computer Animation
MA-INF 2302 – Physics-based Modelling

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 79

MA-INF 2312 Image Acquisition and Analysis in Neuroscience

Workload Credit points Duration Frequency

180 h 6 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Thomas Schultz Prof. Dr. Thomas Schultz
Programme Mode Semester
M. Sc. Computer Science Optional 1-4.

Learning goals: technical skills

Students will learn about image acquisition and analysis pipelines which are used in neuroscience. They will
understand algorithms for image reconstruction, artifact removal, image registration and segmentation, as well as
relevant statistical and machine learning techniques. A particular focus will be on data from Magnetic Resonance
Imaging and on mathematical models for functional and diffusion MRI data.

Learning goals: soft skills

Productive work in small teams, self-dependent solution of practical problems in the area of biomedical image
processing, presentation of solution strategies and implementations, self management, critical reflection of
conclusions drawn from complex experimental data.

Contents
This course covers the full image formation and analysis pipeline that is typically used in biomedical studies, from
image acquisition to image processing and statistical analysis.

Prerequisites
Recommended:
Mathematical background (calculus, linear algebra, statistics); imperative programming.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 3 45 T / 45 S 3 S = independent study
Exercises 1 15 T / 75 S 3

Graded exams
Written exam
Ungraded coursework (required for admission to the exam)
The completion of regularly provided exercise sheets. The work can be done in groups of up to three students. A
total of 50% of the points must be achieved. Each student must present a solution to an exercise in the exercise
sessions once.
Literature
• B. Preim, C. Botha: Visual Computing for Medicine: Theory, Algorithms, and Applications. Morgan Kaufmann,
2014
• R.A. Poldrack, J.A. Mumford, T.E. Nichols: Handbook of Functional MRI Data Analysis. Cambridge University
Press, 2011
• D.K. Jones: Diffusion MRI: Theory, Method, and Applications, Oxford University Press, 2011
Master Computer Science — Universität Bonn 80

MA-INF 2316 Lab Digital Material Appearance

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Matthias Hullin Prof. Dr. Matthias Hullin
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

The students will carry out a practical task (project) in the context of the corresponding area, including test and
documentation of the implemented software/system.

Learning goals: soft skills

Ability to properly present and defend design decisions, to prepare readable documentation of software; skills in
constructively collaborating with others in small teams over a longer period of time; ability to classify ones own
results into the state-of-the-art of the resp. area

Contents
Within the lab, a computational imaging and/or digital material appearance technique is implemented. This can be
a newly developed approach, an improvement of an existing technique, or an implementation of a paper. Each
participant selects a topic, defines an objective, and presents a work plan at the beginning of the project. At the end
of the lab, each project is presented within a seminar and a written report.

Prerequisites
none
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 81

MA-INF 2317 Numerical Algorithms for Visual Computing and Machine

Learning

Workload Credit points Duration Frequency

180 h 6 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Florian Bernard Prof. Dr. Florian Bernard
Programme Mode Semester
M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

• ability to implement basic numerical algorithms, understanding their strengths and shortcomings
• mathematical modelling of computational problems in visual computing and machine learning
• gain an intuition which algorithm is best applied for which problem in visual computing and machine learning, so
that practical problems in these areas can be solved

Learning goals: soft skills

• problem solving skills: ability to identify and utilise analogies between new problems and previously seen ones
• analytical and abstract thinking: develop a general intuition of computational problems, being able to adopt
different perspectives of particular concepts

Contents
This module focuses on numerical methods that frequently occur in the fields visual computing (VC) and machine
learning (ML). In addition to algorithms, this module will also cover modelling aspects that are relevant for solving
practical problems in VC and ML. The contents include:
• Error analysis and conditioning of problems
• Linear systems (solvability, algorithms, stability, regularisation), and applications and modelling in VC and ML
(e.g. linear regression, image alignment, deconvolution)
• Spectral methods (eigenvalue decomposition, singular value decomposition, respective algorithms), and their
applications and modelling in VC and ML (e.g. clustering, Procrustes analysis, point-cloud alignment, principal
components analysis)
• Numerical optimisation (gradient-based methods, second-order methods, large-scale optimisation) and applications
and modelling in VC and ML.

Prerequisites
Recommended:
Participants are expected to have a high level of mathematical maturity (in particular, a good working knowledge of
linear algebra and calculus/analysis is essential). A basic understanding of mathematical optimisation is
advantageous.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to two students. A total
of 50% of the points must be achieved.
Master Computer Science — Universität Bonn 82

MA-INF 2318 Seminar Computational Medical Imaging

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Thomas Schultz Prof. Dr. Shadi Albarqouni

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

• Ability to understand, survey, and critically appraise current research results in computational medical imaging –
including both classical algorithms and deep learning-based methods.
• Deepen one’s grasp of topics such as medical image segmentation, registration, classification, uncertainty
quantification, and bias/causality in medical images.
• Learn to extract the core contributions of a paper, identify strengths and limitations of proposed methods, and
relate them to the state-of-the-art in the field.
Learning goals: soft skills
• Develop competence in independent literature search, paper selection, and reading state-of-the-art scientific
publications.
• Acquire skills to present complex material in a didactically effective manner (both written and oral), including
selecting appropriate visual media (slides, diagrams).
• Practice critical discussion and defense of one’s assessments in a knowledgeable scientific audience.
• Manage one’s time under relatively open assignments and long-ranging deadlines; formulate and accept
constructive criticism.
Contents
Seminar topics cover recent research in computational medical imaging drawn from top journals and conferences.
Typical subject areas include:
• Segmentation & Classification (e.g., anatomical structure delineation, lesion detection)
• Registration & Alignment (e.g., atlas-based, diffeomorphic registration)
• Quantitative Imaging & Radiomics (e.g., feature extraction from images, prognostic modeling)
• Uncertainty Quantification & Bias (e.g., Bayesian deep learning, causal analysis, dataset shift)
• Deep Learning Architectures (e.g., U-Nets, GANs, Vision Transformers)
• Ethical, Regulatory, and Clinical Translation Considerations
Relevant journals:
• Nature Machine Intelligence.
• Medical Image Analysis (MedIA).
• IEEE Transactions on Medical Imaging (TMI).
• European Radiology, Radiology: AI, IEEE Journal of Biomedical and Health Informatics.
Relevant conferences:
• MICCAI (Medical Image Computing & Computer-Assisted Intervention).
• IEEE/CVF Computer Vision and Pattern Recognition (CVPR).
• MIDL (Medical Imaging with Deep Learning).
• ISBI (International Symposium on Biomedical Imaging).
• RSNA (Radiological Society of North America Annual Meeting).

Prerequisites
Recommended:
At least one of the following:
• MA-INF 2222 – Visual Data Analysis.
• MA-INF 2312 – Image Acquisition and Analysis in Neuroscience.
A solid background in programming (preferably Python), as well as basic knowledge of linear algebra, probability,
and numerical algorithms, is required.
Master Computer Science — Universität Bonn 83

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation (20 minutes per student, including Q&A). Written seminar report (8–10 pages summarizing each
student’s topic and critical evaluation).

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 84

3 Information and Communication Management

MA-INF 3108 L2E2 6 CP Secure Software Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 85

MA-INF 3109 L2E2 6 CP Quantum Algorithms: Introduction and Data Fusion Examples . . . . . . . . . . 86
MA-INF 3110 Lab4 9 CP Lab Basics of iOS-Development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
MA-INF 3202 L2E2 6 CP Mobile Communication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 89
MA-INF 3209 Sem2 4 CP Seminar Selected Topics in Communication Management . . . . . . . . . . . . . . . . . 90
MA-INF 3216 Sem2 4 CP Seminar Sensor Data Fusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91
MA-INF 3229 Lab4 9 CP Lab IT-Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 92
MA-INF 3233 L2E2 6 CP Advanced Sensor Data Fusion in Distributed Systems . . . . . . . . . . . . . . . . . . . . 93
MA-INF 3236 L2E2 6 CP IT Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94
MA-INF 3237 L2E2 6 CP Array Signal and Multi-channel Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 96
MA-INF 3238 L2E2 6 CP Side Channel Attacks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97
MA-INF 3239 L2E2 6 CP Malware Analysis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 98
MA-INF 3241 L3E1 6 CP Practical Challenges in Human Factors of Security and Privacy . . . . . . . . . 100
MA-INF 3242 L2E2 6 CP Security of Distributed and Resource-constrained Systems . . . . . . . . . . . . . . 101
MA-INF 3246 L2E2 6 CP Security in Digital Supply Chains . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102
MA-INF 3304 Lab4 9 CP Lab Communication and Communicating Devices . . . . . . . . . . . . . . . . . . . . . . . 103
MA-INF 3310 L2E2 6 CP Introduction to Sensor Data Fusion - Methods and Applications . . . . . . . . 104
MA-INF 3312 Lab4 9 CP Lab Sensor Data Fusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
MA-INF 3317 Sem2 4 CP Seminar Selected Topics in IT Security . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 106
MA-INF 3319 Lab4 9 CP Lab Usable Security and Privacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107
MA-INF 3320 Lab4 9 CP Lab Security in Distributed Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108
MA-INF 3321 Sem2 4 CP Seminar Usable Security and Privacy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 109
MA-INF 3322 L2E2 6 CP Applied Binary Exploitation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 110
Master Computer Science — Universität Bonn 85

MA-INF 3108 Secure Software Engineering

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Dr. Christian Tiefenau Dr. Christian Tiefenau
Programme Mode Semester
M. Sc. Computer Science Optional 1-3.

Learning goals: technical skills

The students will learn how to integrate security aspects into the phases of a Software Development Lifecycle. They
will learn:
• Methods for identifying threats and vulnerabilities (threat modeling and risk analysis).
• How to design secure architectures using fundamental design principles (e.g., STRIDE).
• Secure coding practices and common vulnerability types.
• Key considerations when using cryptographic methods in software.
• How to assess the severity of a vulnerability.
• Best practices for system configuration, deployment, and maintenance.

Learning goals: soft skills

In the exercises, the students will conduct practical tasks to strengthen the understanding of the methods within the
secure software engineering lifecycle. Through this, the abilities teamwork, organization and critical discussion of
their own and others’ results are strengthened.

Contents
• Threat modeling
• Risk analysis
• Architectural security
• Secure coding
• Applied Cryptography
• Secure configuration and deployment
• Updates and maintenance

Prerequisites
Recommended:
Basic knowledge of software engineering and IT Security-concepts is advantageous but not mandatory.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (90 minutes)

Ungraded coursework (required for admission to the exam)

none
Literature
Software Security: Building Security In by Gary McGraw
Master Computer Science — Universität Bonn 86

MA-INF 3109 Quantum Algorithms: Introduction and Data Fusion

Examples

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Wolfgang Koch Prof. Dr. Wolfgang Koch, Dr. Felix Govaers, Dr. Martin Ulmke

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Quantum algorithms for data fusion may become game changers as soon as quantum processing kernels embedded in
hybrid processing architectures with classical processors will exist. While emerging quantum technologies directly
apply quantum physics, quantum algorithms do not exploit quantum physical phenomena as such, but rather use the
sophisticated framework of quantum physics to deal with “uncertainty”. Although the link between mathematical
statistics and quantum physics has long been known, the potential of physics-inspired algorithms for data fusion has
just begun to be realized. While the implementation of quantum algorithms is to be considered on classical as well as
on quantum computers, the latter are anticipated as well-adapted “analog computers” for unprecedentedly fast
solving data fusion and resources management problems. While the development of quantum computers cannot be
taken for granted, their potential is nonetheless real and has to be considered by the international information fusion
community.

Learning goals: soft skills

• Problem solving
• Adaptability
• Critical thinking

Contents
• Introduction with Examples
• Short introduction to quantum mechanics
• Introduction to quantum computing
• Quantum computing hardware
• Quantum inspired tracking
• Particle filtering and fermionic target tracking
• The data association problem
• Track extraction and sensor management
• Quantum computing for multi target tracking data association
• Quantum computing for resources management
• Quantum many particle systems and boson sampling
• Path Integrals

Prerequisites
Recommended:
One of the following:
• BA-INF 137 – Einführung in die Sensordatenfusion
• MA-INF 3310 – Introduction to Sensor Data Fusion - Methods and Applications

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam
Master Computer Science — Universität Bonn 87

Ungraded coursework (required for admission to the exam)

50% of the possible points for the exercises. The points are acquired by a small programming exercise with a
workload of about 15 hours and some theoretical exercises with a workload of 10 hours. The solution has to be
submitted individually or in groups of up to three students and will be rated by points.
Master Computer Science — Universität Bonn 88

MA-INF 3110 Lab Basics of iOS-Development

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Stephan Jonas Prof. Dr. Stephan Jonas

Programme Mode Semester

M. Sc. Computer Science Optional 1-3.

Learning goals: technical skills

Students acquire the ability to design and develop iOS applications while critically evaluating current research and
best practices in mobile computing. They learn to compare different frameworks and architectural approaches,
justify their decisions based on literature, and reflect on the advantages and limitations of various solutions. In
addition to implementation, emphasis is placed on critical analysis and integration of current research findings, e.g.,
in the areas of usability, accessibility, performance, and security.
Students learn how to conduct literature research related to digital medicine application development and to
critically evaluate scientific sources in mobile app development and human-computer interaction. They learn to
document and justify design and development decisions with reference to scientific findings and empirical evidence.
Furthermore, students learn to communicate results in a scientifically grounded project report and oral presentation.

Learning goals: soft skills

• Self-competences: Work and time management, handling feedback, and adapting to changing requirements.
• Develop scientific thinking and research literacy by critically evaluating sources, documenting decisions, and
reflecting on the app development process in an evidence-based manner.

Contents
• Usage of Xcode
• Basic understanding of the programming language Swift.
• UI development with SwiftUI.
• Design of User Interfaces and Prototyping.
• Local data persistence on the devices.
• Git: Branches, Pull Requests, Code Review.

Prerequisites
Recommended:
Experience in object-oriented programming

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation (individually graded), written project documentation / report.

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Forms of media
Projection, videos, interactive exercises.
The lectures and exercises are based on interaction. We expect active participation! To take part in the exercises, a
MacBook (with macOS 15 or later) is required. Please bring your own device or contact us in advance if you need to
borrow one.
Literature
• Human Interface Guidelines ([Link]
• SwiftUI Dokumentation ([Link]
Master Computer Science — Universität Bonn 89

MA-INF 3202 Mobile Communication

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Peter Martini Prof. Dr. Peter Martini, Dr. Matthias Frank

Programme Mode Semester

M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

After completion of the module students will be able to cope with challenges and problems arising in design and
operation of wireless and mobile communication systems. They can choose suitable protocols or design new ones.
They are able to select mechanisms from different architectural layers, integrate them into a new complete
architecture and justify their selection and integration decisions.

Learning goals: soft skills

Theoretical exercises support in-depth understanding of lecture topics and stimulate discussions; practical exercises
in teamwork support time management, targeted organisation of practical work and critical discussion of own and
others’ results
Contents
Mobility Management in the Internet, Wireless Communication Basics, Wireless Networking Technologies (like
Bluetooth, Wireless LAN, LoRa/LoRaWAN, focus on system architecture and medium access), Cellular/Mobile
Communication Networks (voice and data communication, 2G, 4G, ...).

Prerequisites
Recommended:
Bachelor level knowledge of basics of communication systems and Internet protocols. Students may receive access to
lecture slides in English language of our Bachelor module BA-INF 101 "Kommunikation in Verteilten Systemen" as a
reference. Contact the lecturer in advance of the course, and information will also be given in the first lecture.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (90 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to three students. A
total of 50% of the points must be achieved. For 70% of the exercise sheets, 20% of the points must be achieved for
each sheet.
Literature
• Jochen Schiller: Mobile Communications, Addison-Wesley, 2003
• William Stallings: Wireless Communications and Networking, Prentice Hall, 2002
• Further up-to-date literature will be announced in due course before the beginning of the lecture
Master Computer Science — Universität Bonn 90

MA-INF 3209 Seminar Selected Topics in Communication Management

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Peter Martini Prof. Dr. Peter Martini, Prof. Dr. Michael Meier, Dr. Matthias Frank

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Enhanced and in-depth knowledge in specialized topics in the area of communication systems and Internet protocols.

Learning goals: soft skills

Acquire the competence to independently search for and study state-of-the-art scientific literature in depth, read
critically, identify the most relevant content, and assess research results in the context of the corresponding research
area; to discuss research results with a knowledgeable scientific audience; to present prior work by others in writing
and in presentations with visual media in a way that adheres to academic standards, that is well-structured and
didactically effective, and that motivates the audience to participate; to formulate and accept criticism; to manage
one’s time with relatively open assignments and long-ranging deadlines.

Contents
Current research topics in communication systems and Internet protocols based on conference and journal papers.

Prerequisites
Recommended:
Successful completion of at least one of the following lectures:
• MA-INF 3202 Mobile Communication
• MA-INF 3236 IT Security
• MA-INF 3239 Malware Analysis
Bachelor level knowledge of basics of communication systems and Internet protocols, e.g. OSI model, medium access
of wired and wireless LAN technologies, IP adressing and routing, transport protocols UDP and TCP.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
The relevant literature will be announced towards the end of the previous semester
Master Computer Science — Universität Bonn 91

MA-INF 3216 Seminar Sensor Data Fusion

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Wolfgang Koch Prof. Dr. Wolfgang Koch, Dr. Felix Govaers

Programme Mode Semester

M. Sc. Computer Science Optional 3.

Learning goals: technical skills

Enhanced and in-depth knowledge in specialized topics in state estimation and object tracking.

Learning goals: soft skills

Acquire the competence to independently search for and study state-of-the-art scientific literature in depth, read
critically, identify the most relevant content, and assess research results in the context of the corresponding research
area; to discuss research results with a knowledgeable scientific audience; to present prior work by others in writing
and in presentations with visual media in a way that adheres to academic standards, that is well-structured and
didactically effective, and that motivates the audience to participate; to formulate and accept criticism; to manage
one’s time with relatively open assignments and long-ranging deadlines.

Contents
The seminar focuses on specific research papers in the field of sensor data fusion, which may include topics like
non-linear state estimation, deep learning for sensor perception, or multi object tracking.

Prerequisites
Recommended:
• MA-INF 3310 Introduction to Sensor Data Fusion – Methods and Applications.
• It is assumed that the participants know linear algebra and have basic knowledge in probability theory.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
The relevant literature will be announced at the beginning of the seminar.
Master Computer Science — Universität Bonn 92

MA-INF 3229 Lab IT-Security

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Michael Meier Prof. Dr. Michael Meier
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

After completion of the lab module students have completed a practical task in the context of IT security. The
students will have gained experience in the typical technical skills like the design and implementation of software,
test and documentation of the software, and performance evaluation (e.g. by measurements, simulation, analysis)
and presentation of performance results.

Learning goals: soft skills

Ability to analyze problems theoretically and to find creative and practical solutions; to examine one’s solutions and
results critically; to classify one’s own results into the state-of-the-art of the respective area; to prepare readable
documentation of software and research results; to present, defend and discuss design decisions and results in the
team/group and to other students clearly and in accordance with academic standards; to collaborate constructively
with others in small teams over a longer period of time; to aim at long-range goals under limited resources; to work
under pressure.

Contents
Selected topics close to current research in the area of IT security.

Prerequisites
Recommended:
Foundational knowledge in
• IT security: security terminology, authentication, access control, applied cryptography (symmetric encryption,
asymmetric encryption, hashing, key management)
• Networks: OSI model, addressing, routing, protocols.
It is recommended to take MA-INF 3236 IT Security first.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 93

MA-INF 3233 Advanced Sensor Data Fusion in Distributed Systems

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Wolfgang Koch Dr. Felix Govaers

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Students will be able to describe the advanced principles of sensor data fusion for state estimation and object tracking
based on tracks from multiple sensors in distributed systems. They will be aware of the correlation problem in
track-to-track fusion and know several algorithms to cope with it. They will know the assumptions, advantages and
disadvantages of different algorithms and be able to select and apply suitable candidates in practical applications.

Learning goals: soft skills

Mathematical derivation of algorithms, application of mathematical results on estimation theory.

Contents
Tracklet fusion, the Bar-Shalom-Campo formula, the Federated Kalman Filter, naive fusion, the distributed Kalman
filter and the least squares estimate, Accumulated State Densities, Decorrlated fusion, product representation.
For challenging state estimation tasks, algorithms which enhance the situational awareness by fusing sensor
information are inevitable. Nowadays it has become very popular to improve the performance of systems by linking
multiple sensors. This implies some challenges to the sensor data fusion methodologies such as sensor registration,
communication delays, and correlations of estimation errors. In particular, if the communication links have limited
bandwidth, data reduction techniques have to be applied at the sensor sites, that is local tracks have to be
computed. Once recieved at a fusion center (FC), the tracks then are fused to reconstruct a global estimate.

Prerequisites
Recommended:
Basic knowledge about the Kalman filter is required (see also recommended literature).
Students who did not take BA-INF 137 – Einführung in die Sensordatenfusion in their Bachelor’s are advised to first
take MA-INF 3310 – Introduction to Sensor Data Fusion - Methods and Applications.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam (30 minutes)

Ungraded coursework (required for admission to the exam)

50% of the maximum achievable points in the practical programming exercises are required. There is one practical
exercise, which is a workload of about 15h. The delivery of the programmed solution is done individually or in group
work of up to three students. A total of 10 points will be awarded, 50% of which will have been achieved if the
Distributed Kalman filter has been programmed in an executable and consistent manner.

Forms of media
Power Point
Literature
W. Koch: "Tracking and Sensor Data Fusion: Methodological Framework and Selected Applications", Springer, 2014.
D. Hall, C.-Y. Chong, J. Llinas, and M. L. II: "Distributed Data Fusion for Network-Centric Operations", CRC
Press, 2014.
Master Computer Science — Universität Bonn 94

MA-INF 3236 IT Security

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Michael Meier Prof. Dr. Michael Meier
Programme Mode Semester
M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

Knowledge of a variety of active research fields in IT security including motivation, challenges and objectives in these
fields as well as selected fundamental knowledge and methods helping students to deepen their knowledge in their
upcoming studies. In detail, participants will know
• advanced cryptographic constructions and low-level programming in offensive and defensive scenarios;
• how to apply program analysis techniques to IT security;
• how to achieve security by and security of methods from the area of machine learning.

Learning goals: soft skills

Working in small groups on exercises, critical discussion of own and others’ results, time management, transferring
theoretical knowledge to practical scenarios

Contents
The contents vary but usually include
• Privacy
• Cryptographic Protocols
• Network Security
• Supply Chain Attacks
• Management of Identity Data
• Low-level software analysis
• Software testing
• Side Channel Attacks
• Anomaly Detection
• Human Factor in Security

Prerequisites
Recommended:
Foundational knowledge in
• IT security: security terminology, authentication, access control, applied cryptography (symmetric encryption,
asymmetric encryption, hashing, key management)
• Low-level/OS-level programming: x86 assembly, C programming, OS-level programming for Linux, buffer
overflows, sockets
• Networking: OSI model, modulation, addressing, routing, udp, tcp
You find useful information on these topics in the following books (available through library search portal bonnus):
• M. Bishop: Computer Security: Art and Science, Pearson Education, 2018.
• J. Streib: Guide to Assembly Language: A Concise Introduction. Springer, 2020.
• W. Stevens: UNIX Network Programming – The Sockets Networking API, Prentice Hall International, 3rd
Edition, 2003
• Tanenbaum: Computer Networks, Pearson Education, 4th Edition, 2002

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 minutes)
Master Computer Science — Universität Bonn 95

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to three students. For
70% of the exercise sheets, 50% of the points must be achieved for each sheet.
Master Computer Science — Universität Bonn 96

MA-INF 3237 Array Signal and Multi-channel Processing

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Wolfgang Koch Dr. Marc Oispuu

Programme Mode Semester

M. Sc. Computer Science Optional 1-3.

Learning goals: technical skills

Students will be able to describe the central principles of array signal and multi-channel processing and to name
their advantages and disadvantages as well as to illustrate their relevance in application examples such as wireless
communication, acoustics, radar, sonar or seismology. Furthermore, they will be able to implement suitable methods
of direction finding, spatial filtering and bearings-only localization and to apply them to electromagnetic or acoustic
signals and to evaluate the achieved results in terms of their performance.

Learning goals: soft skills

Mathematical derivation of algorithms, applications of mathematical results on estimation theory

Contents
Estimation theory, Sensor model, Cramér-Rao analysis, conventional beamforming, Multiple Signal Classification
(MUSIC), sensor calibration, Bearings-only localization, Direct Position Determination (DPD), Applications

Prerequisites
none
Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral Exam (30 minutes)

Ungraded coursework (required for admission to the exam)

50% of the maximum achievable points in the practical programming exercises are required. There is one practical
exercise, which is a workload of about 10h. The delivery of the programmed solution is done individually or in group
work of up to three students. A total of 10 points will be awarded, 50% of which will have been achieved if the basic
signal processing algorithms for array sensors have been implemented.

Forms of media
Power Point
Literature
H. L. van Trees, Optimum Array Processing. Part IV of Detection, Estimation, and Modulation Theory. New York:
Wiley-Interscience, 2002.
Master Computer Science — Universität Bonn 97

MA-INF 3238 Side Channel Attacks

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Dr. Felix Boes Dr. Felix Boes
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

• Students are introduced to theoretical and practical side channel effects of modern hardware.
• Students learn techniques to utilize these effects to circumvent security mechanisms.
• This includes covert channels as well as side channel attacks and microarchitectural attacks on modern CPUs.
Learning goals: soft skills
Theoretical exercises to support in-depth understanding of lecture topics and to stimulate discussions, practical
exercises in teamwork to support time management, targeted organization of practical work and critical discussion of
own and others’ results.
Contents
• Theoretical foundations of side channel effects and attacks as well as
• covert channels,
• differential power analysis,
• padding oracle,
• RSA timing attacks,
• cache based side channel effects,
• microarchitectural attacks (Spectre)

Prerequisites
Recommended:
Fundamental knowledge about IT Security, operating systems and statistics is advantageous but not mandatory.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam
Ungraded coursework (required for admission to the exam)
Participation in two achievement tests. In total, at least 50% of the points much be achieved on these tests.
Master Computer Science — Universität Bonn 98

MA-INF 3239 Malware Analysis

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Peter Martini Prof. Dr. Elmar Padilla
Programme Mode Semester
M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

The students should be able to analyze the functional scope of a binary file independently and to describe its damage
potential. In addition, the students should be able to carry out detailed analyzes of given aspects and to partially
automate these with the help of scripts.

Learning goals: soft skills

Presentation of solutions and methods, critical discussion of applied methods and techniques.

Contents
In the course, the skills acquired so far in binary analysis will first be deepened and adapted to the peculiarities of
malware analysis. Different malware samples are used to explain the techniques used by malware authors. These
priorities include:
• Characteristics of malware
• Persistence
• Network communication
• Encryption
• Dynamic malware analysis
• Debugging
• Behavioral obfuscation
• Virtual analysis environments
• Static malware analysis
• Control flow obfuscation
• Automation of common analysis steps
• Reconstruction of binary algorithms
The event begins with several lectures that provide the basics for the students to work independently later. In the
course of this, the students will work on practical topics from the field of malware analysis during the semester.
Since these subject areas can turn out to be very specific, it is necessary to be willing to deal with the subject
outside of the lecture and exercise times.
Prerequisites
Recommended:
Basic knowledge of operating systems (kernel, threads, virtual memory), network communication (protocols,
architectures), binary analysis (assembler, endianness, semantic gap, coding), software development (programming,
semantics, scripting in Python)

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam (30 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to two students. A total
of 50% of the points must be achieved.
Master Computer Science — Universität Bonn 99

Literature
The relevant literature will be announced at the beginning of the lecture
Master Computer Science — Universität Bonn 100

MA-INF 3241 Practical Challenges in Human Factors of Security and

Privacy

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Matthew Smith Prof. Dr. Matthew Smith
Programme Mode Semester
M. Sc. Computer Science Optional 1-3.

Learning goals: technical skills

Students are introduced to a variety of current challenges in security and privacy which contain human aspects and
have societal relevance. Students learn about the motivation, challenges and objectives in these areas and select one
for the semester topic. They then learn how to design, conduct and evaluate user studies to tackle the selected
challenge. Additionally, they get to know selected fundamental knowledge and methods helping them to deepen their
knowledge on human factors research.

Learning goals: soft skills

Breaking down complex topics into manageable components, critical discussion of own and others’ results, time
management, transferring theoretical knowledge to practical scenarios

Contents
The course begins with several lectures that provide an overview and discussion of current societal challenges in the
area of human factors in security and privacy. The students will select a semester topic and together with the
lecturer explore this topic using user studies. Since these subject areas can turn out to be very specific, it is
beneficial to be willing to deal with the subject outside of the lecture and exercise times. Topics can include
surveillance, age verification, anonymity, online abuse, fake news, etc.

Prerequisites
Recommended:
It is recommended that students have experience with designing and evaluating survey and interview-based user
studies. It is recommended to check the material of BA-INF 145 Usable Security and Privacy (available in English).

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 1 15 T / 45 S 2 S = independent study
Exercises 3 45 T / 75 S 4

Graded exams
electronic exam (90 minutes, pass/fail)

Ungraded coursework (required for admission to the exam)

The participation in at least 80% of regularly provided exercises.
Master Computer Science — Universität Bonn 101

MA-INF 3242 Security of Distributed and Resource-constrained Systems

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Michael Meier Dr. Thorsten Aurisch
Programme Mode Semester
M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

Ability to understand and analyse theoretical and practical cyber security challenges of distributed and
ressource-constrained systems, as well as the ability to select and apply appropriate solutions.

Learning goals: soft skills

Working in small groups on exercises, critical discussion of own and others’ results, time management, transferring
theoretical knowledge to practical scenarios.

Contents
• Group communication with IP multicast
• Group key management
• Broadcast encryption
• Public key infrastructure
• Web of trust
• Multicast infrastructure protection
• Distributed security mechanisms
• Cyber resilience in groups
• Distributed ledger technology
• Cyber security in software-defined networks
• Artificial intelligence in cyber security
• Security for IoT

Prerequisites
Recommended:
Foundational knowledge in
• IT security: security terminology, authentication, access control, applied cryptography (symmetric encryption,
asymmetric encryption, hashing, key management)
• Low-level/OS-level programming: x86 assembly, C programming, OS-level programming for Linux, buffer
overflows, sockets
• Networking: OSI model, modulation, addressing, routing, udp, tcp
You find useful information on these topics in the following books (available through library search portal bonnus):
• M. Bishop: Computer Security: Art and Science, Pearson Education, 2018.
• J. Streib: Guide to Assembly Language: A Concise Introduction. Springer, 2020.
• W. Stevens: UNIX Network Programming – The Sockets Networking API, Prentice Hall International, 3rd
Edition, 2003
• Tanenbaum: Computer Networks, Pearson Education, 4th Edition, 2002

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 minutes)

Ungraded coursework (required for admission to the exam)

Participation in an achievement test. At least 50% of the points much be achieved on this test.
Master Computer Science — Universität Bonn 102

MA-INF 3246 Security in Digital Supply Chains

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Dr. Marc Ohm Dr. Marc Ohm
Programme Mode Semester
M. Sc. Computer Science Optional 2-3.

Learning goals: technical skills

This module introduces the challenges and risks of digital supply chains in the context of cybersecurity. It focuses on
recent developments in the software supply chain and the artificial intelligence supply chain. Additionally, it will
present threat intelligence methodologies.

Learning goals: soft skills

Presentation of solutions and methods, critical discussion of own and others’ results.

Contents
• Threat Actors
• Threat Intelligence
• Attack vector of Software Supply Chains
• Adversarial Machine Learning
• Prevention and mitigation strategies
• Regulations and compliance

Prerequisites
Recommended:
• MA-INF 3236 IT-Security
• MA-INF 4204 Technical Neural Nets
An understanding of the basic concepts of software development, artificial intelligence and IT security.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam
Ungraded coursework (required for admission to the exam)
Participation in a performance test. At least 50% of the points much be achieved on this test.
Master Computer Science — Universität Bonn 103

MA-INF 3304 Lab Communication and Communicating Devices

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Peter Martini Prof. Dr. Peter Martini, Prof. Dr. Michael Meier, Dr. Matthias Frank

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

After completion of the lab module students have completed a practical task in the context of communication and
networked systems. The students will have gained experience in the typical technical skills like the design and
implementation of communication software/networked systems, test and documentation of the software, and
performance evaluation (e.g. by measurements, simulation, analysis) and presentation of performance results.

Learning goals: soft skills

Ability to analyze problems theoretically and to find creative and practical solutions; to examine one’s solutions and
results critically; to classify one’s own results into the state-of-the-art of the respective area; to prepare readable
documentation of software and research results; to present, defend and discuss design decisions and results in the
team/group and to other students clearly and in accordance with academic standards; to collaborate constructively
with others in small teams over a longer period of time; to aim at long-range goals under limited resources; to work
under pressure.

Contents
Selected topics close to current research in the area of communication systems, network security, mobile
communication and communicating devices.

Prerequisites
Recommended:
Foundational knowledge in networks: OSI model, addressing, routing, protocols;
Successful completion of at least one of the following lectures:
• MA-INF 3202 Mobile Communication
• MA-INF 3236 IT Security
• MA-INF 3239 Malware Analysis

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
The relevant literature will be announced towards the end of the previous semester.
Master Computer Science — Universität Bonn 104

MA-INF 3310 Introduction to Sensor Data Fusion - Methods and

Applications

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Wolfgang Koch Prof. Dr. Wolfgang Koch

Programme Mode Semester

M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

Students will be able to describe the central principles of sensor data fusion for state estimation and object tracking
based on error-prone and ambiguous measurements. They will be able to apply the Kalman filter and to formulate
sensor and dynamics models for different kind of sensors and objects. Furthermore, they will know the concept of
probabilistic data association and a computationally feasible approximation using Gaussian mixture densities.

Learning goals: soft skills

Mathematical derivation of algorithms, application of mathematical results on estimation theory.

Contents
Gaussian probability density functions, Kalman filter, Unscented Kalman Filter, Extended Kalman Filter, Particle
Filter, Multi-Hypothesis-Trackier, Extended Target Tracking, Road Tracking, Interacting Multiple Model Filter,
Retrodiction, Smoothing, Maneuver Modeling.
The lecture starts with preliminaries on how to handle uncertain data and knowledge within analytical calculus.
Then, the fundamental and well-known Kalman filter is derived. Based on this tracking scheme, further approaches
to a wide spectrum of applications will be shown. All algorithms will be motivated by examples from ongoing
research projects, industrial cooperations, and impressions of current demonstration hardware.
Because of inherent practical issues, every sensor measures certain properties up to an error. This lecture shows how
to model and overcome this error by an application of theoretical tools such as Bayes’ rule and further derivations.
Moreover, solutions to possible false-alarms, miss-detections, maneuvering phases, and much more will be presented.

Prerequisites
Recommended:
It is assumed that the participants know linear algebra and have basic knowledge in probability theory.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral exam
Ungraded coursework (required for admission to the exam)
50% of the possible points for the exercises. The points are acquired by a small programming exercise with a
workload of about 15 hours and some theoretical exercises with a workload of 10 hours. The solution has to be
submitted individually or in groups of up to three students and will be rated by points.

Literature
• W. Koch: "Tracking and Sensor Data Fusion: Methodological Framework and Selected Applications", Springer,
2014.
• Y. Bar-Shalom: "Estimation with Applications to Tracking and Navigation", Wiley-Interscience, 2001.
Master Computer Science — Universität Bonn 105

MA-INF 3312 Lab Sensor Data Fusion

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Wolfgang Koch Prof. Dr. Wolfgang Koch

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Enhanced and in-depth knowledge in specialized topics in state estimation and object tracking.

Learning goals: soft skills

Ability to analyze problems theoretically and to find creative and practical solutions; to examine one’s solutions and
results critically; to classify one’s own results into the state-of-the-art of the respective area; to prepare readable
documentation of software and research results; to present, defend and discuss design decisions and results in the
team/group and to other students clearly and in accordance with academic standards; to collaborate constructively
with others in small teams over a longer period of time; to aim at long-range goals under limited resources; to work
under pressure.

Contents
In the lab, the students apply methods from the sensor data fusion and state estimation theory to practical examples
in order to get experience in the application and implementation. The exemplary scenarios and application examples
vary each year but may be for instance on a simulated radar network for multi object tracking, camera image
processing, heterogeneous sensor fusion, or array sensor bearing processing.
The students shall work together in a team. Everyone is responsible for a specific part in the context of a main goal.
Results will be exchanged and integrated via software interfaces.

Prerequisites
Recommended:
• MA-INF 3310 Introduction to Sensor Data Fusion – Methods and Applications;
• knowledge of linear algebra and basic knowledge in probability theory.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
The relevant literature will be announced at the beginning of the lab.
Master Computer Science — Universität Bonn 106

MA-INF 3317 Seminar Selected Topics in IT Security

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Michael Meier Prof. Dr. Michael Meier, Prof. Dr. Peter Martini

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Enhanced and in-depth knowledge in specialized topics in the area of IT Security.

Learning goals: soft skills

Acquire the competence to independently search for and study state-of-the-art scientific literature in depth, read
critically, identify the most relevant content, and assess research results in the context of the corresponding research
area; to discuss research results with a knowledgeable scientific audience; to present prior work by others in writing
and in presentations with visual media in a way that adheres to academic standards, that is well-structured and
didactically effective, and that motivates the audience to participate; to formulate and accept criticism; to manage
one’s time with relatively open assignments and long-ranging deadlines.

Contents
Current research topics in IT security based on conference and journal papers.

Prerequisites
Recommended:
Foundational knowledge in
• IT security: security terminology, authentication, access control, applied cryptography (symmetric encryption,
asymmetric encryption, hashing, key management)
• Networks: OSI model, addressing, routing, protocols.
It is recommended to take MA-INF 3236 IT Security first.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 107

MA-INF 3319 Lab Usable Security and Privacy

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Matthew Smith Prof. Dr. Matthew Smith
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Ability to carry out a practical task (project) in the context of human factors of security and privacy, including user
studies and their evaluation. This includes selecting variables of interest, designing measurement instruments, such
as interviews, surveys or prototypes, recruiting participants, executing and evaluating the user-study.

Learning goals: soft skills

Ability to work in small teams and cooperate with other teams in a group; ability to make design decisions in a
practical task; present and discuss (interim and final) results in the team/group and to other students; prepare
written documentation of the work carried out
Contents
The students will select and carry out a practical task (project) in the context of human factors of security and
privacy, including user studies and their evaluation. Topics for the project are close to current research in the area of
human aspects of security and privacy. Focus topics include but are not limited to: Attitudes towards Surveillance,
S&P Ethics, Privacy technology, Authentication, Encryption, Gamification, Age verification, etc.

Prerequisites
Recommended:
Knowledge on how to run and evaluate user studies is required. It is recommended to check the material of the
Bachelor’s course BA-INF 145 Usable Security and Privacy (available in English) and to take:
• MA-INF 3241 Practical Challenges in Human Factors of Security and Privacy.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 108

MA-INF 3320 Lab Security in Distributed Systems

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Matthew Smith Prof. Dr. Matthew Smith
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Ability to carry out a practical task (project) in the context of distributed security using modern software
engineering processes, including testing and documentation of the implemented software/system.

Learning goals: soft skills

Ability to work in small teams and cooperate with other teams in a group; ability to make design decisions in a
practical task; present and discuss (interim and final) results in the team/group and to other students; prepare
written documentation of the work carried out.
Contents
The students will carry out a practical task (project) in the context of distributed security, including documentation
of the implemented software/system. Topics are selected topics close to current research in the area of distributed
systems security and privacy. Focus topics include but are not limited to: Authentication, Encryption, Gamification,
Age verification, Data management, Study platforms, etc. The students will build software systems using modern
software engineering processes. They will test them either programmatically or with a small user studies. They will
document their software.
Prerequisites
Recommended:
Strong programming skills are required. It is recommended to take MA-INF 3242 Security of Distributed and
Resource-constrainted Systems first.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 109

MA-INF 3321 Seminar Usable Security and Privacy

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Matthew Smith Prof. Dr. Matthew Smith
Programme Mode Semester
M. Sc. Computer Science Optional 1-3.

Learning goals: technical skills

Enhanced and in-depth knowledge in specialized topics in the area of human aspects of security and privacy.

Learning goals: soft skills

Acquire the competence to independently search for and study state-of-the-art scientific literature in depth, read
critically, identify the most relevant content, and assess research results in the context of the corresponding research
area; to discuss research results with a knowledgeable scientific audience; to present prior work by others in writing
and in presentations with visual media in a way that adheres to academic standards, that is well-structured and
didactically effective, and that motivates the audience to participate; to formulate and accept criticism; to manage
one’s time with relatively open assignments and long-ranging deadlines.

Contents
Current conference and journal papers in the area of human aspects of security and privacy. This includes but is not
limited to: Attitudes towards Surveillance, S&P Ethics, Privacy technology, Authentication, Encryption,
Gamification, Age verification, etc.

Prerequisites
none
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 110

MA-INF 3322 Applied Binary Exploitation

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Peter Martini Prof. Dr. Elmar Padilla
Programme Mode Semester
M. Sc. Computer Science Optional 1-3.

Learning goals: technical skills

recognition of vulnerabilities in binary programs, static reverse engineering of binary programs (Ghidra, IDA Free,
Linux command line tools), debugging of binary programs with gdb/pwndbg, Python programming with pwntools,
application of exploit strategies such as overwriting return addresses/function pointers, return-oriented programming
(ROP, SROP, ret2csu), shellcoding, glibc heap exploitation techniques (Use-After-Free, Unlink Exploit, House of
Orange), understanding a complex real-world exploit, usage of git/GitLab and Docker for the exercises.

Learning goals: soft skills

Frustration tolerance when working with binary representations and trying to apply taught techniques, focused
working on technically challenging problems, simultaneously applying knowledge from different areas of computer
science.
Contents
This university course covers various topics related to software security and exploitation techniques. It starts with an
introduction to finding vulnerabilities in C programs and binaries. The course then delves into stack-based buffer
overflows and the mitigations used to prevent them. Students will also learn about circumventing these mitigations
and explore return-oriented programming. The course continues with a focus on manual crafting of shellcode and
understanding the internals of the glibc heap. Students will gain knowledge about heap exploitation techniques,
including use-after-free exploits, heap unlink exploits, and the house of orange exploit. The course concludes with a
complex case study on the Exim RCE exploit, providing students with a practical understanding of real-world
vulnerabilities. Additionally, guest lectures will be held to provide further insights into the field of software security.
Please note that basic skills in static and dynamic binary analysis (e. g. read disassembled/decompiled code or
debug a binary program with gdb) are required to successfully participate in this lecture.

Prerequisites
Recommended:
• Binary Analysis skills (as taught in the Bachelor’s module BA-INF 155 Angewandte Binäranalyse; English slides
available)
• Basic knowledge of the Linux operating system
• System Programming skills in C
• Basic Python programming skills
This module is best taken after or together with MA-INF 3239 Malware Analysis.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Oral Examination (30 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to three students. The
exercises are divided into group tasks and tasks to be completed individually. For each category of tasks, at least
50% of the points must be achieved.

Literature
The relevant literature will be announced at the beginning of the lecture
Master Computer Science — Universität Bonn 111

4 Intelligent Systems

MA-INF 4111 L2E2 6 CP Principles of Machine Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112

MA-INF 4112 L2E2 6 CP Algorithms for Data Science . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114
MA-INF 4113 L2E2 6 CP Cognitive Robotics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115
MA-INF 4114 L2E2 6 CP Robot Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116
MA-INF 4115 L3E1 6 CP Introduction to Natural Language Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 117
MA-INF 4116 Sem2 4 CP Seminar AI Safety . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119
MA-INF 4117 L2E2 4 CP Mining Media Data I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 120
MA-INF 4201 L2E2 6 CP Artificial Life . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121
MA-INF 4204 L2E2 6 CP Technical Neural Nets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 122
MA-INF 4208 Sem2 4 CP Seminar Vision Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124
MA-INF 4209 Sem2 4 CP Seminar Principles of Data Mining and Learning Algorithms . . . . . . . . . . . . . . . . . . . . . . 125
MA-INF 4211 Sem2 4 CP Seminar Cognitive Robotics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 126
MA-INF 4213 Sem2 4 CP Seminar Humanoid Robots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
MA-INF 4214 Lab4 9 CP Lab Humanoid Robots . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 128
MA-INF 4215 L2E2 6 CP Humanoid Robotics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129
MA-INF 4216 L2E2 6 CP Biomedical Data Science and AI . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 130
MA-INF 4217 Sem2 4 CP Seminar Machine Learning Methods in the Life Sciences . . . . . . . . . . . . . . . . . . . . . . . . . . 131
MA-INF 4226 Lab4 9 CP Lab Parallel Computing for Mobile Robotics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 132
MA-INF 4228 L4E2 9 CP Foundations of Data Science . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
MA-INF 4230 L2E2 6 CP Advanced Methods of Information Retrieval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
MA-INF 4231 Sem2 4 CP Seminar Advanced Topics in Information Retrieval . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
MA-INF 4232 Lab4 9 CP Lab Information Retrieval in Practice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
MA-INF 4235 L2E2 6 CP Reinforcement Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
MA-INF 4237 Lab4 9 CP Lab Natural Language Processing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
MA-INF 4240 Lab4 9 CP Lab Hybrid Learning and Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
MA-INF 4241 Lab4 9 CP Lab Cognitive Modelling of Biological Agents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
MA-INF 4242 L2E2 6 CP Self-supervised Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 144
MA-INF 4243 L2E2 4 CP Mining Media Data II . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 145
MA-INF 4244 Lab4 9 CP Lab Deep Learning for the Physical Sciences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 146
MA-INF 4245 Lab4 9 CP Lab AI for Scientific Discovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147
MA-INF 4304 Lab4 9 CP Lab Cognitive Robotics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148
MA-INF 4306 Lab4 9 CP Lab Development and Application of Data Mining and Learning Systems . . . . . . . . . 149
MA-INF 4308 Lab4 9 CP Lab Vision Systems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 150
MA-INF 4322 Lab4 9 CP Lab Machine Learning on Encrypted Data . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151
MA-INF 4324 Sem2 4 CP Seminar Advanced Topics in Data Science . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152
MA-INF 4325 Lab4 9 CP Lab Data Science in Practice . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 153
MA-INF 4326 L2E2 6 CP Explainable AI and Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 154
MA-INF 4327 Lab4 9 CP Lab Biomedical Data Science . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156
MA-INF 4328 L2E2 6 CP Spatio-Temporal Data Analytics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 157
MA-INF 4329 Sem2 4 CP Seminar Biological Intelligence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 158
MA-INF 4330 Lab4 9 CP Lab Explainable AI and Applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159
MA-INF 4331 Lab4 9 CP Lab Perception and Learning for Robotics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 161
MA-INF 4332 Sem2 4 CP Seminar Large Language Models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162
MA-INF 4333 L2E2 6 CP Geometric Deep Learning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 163
MA-INF 4334 L2E2 6 CP Computational neuroscience: cognition and behaviour . . . . . . . . . . . . . . . . . . . . . . . . . . . . 164
MA-INF 4335 Lab4 9 CP Lab AI Alignment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165
MA-INF 4336 Sem2 4 CP Seminar Selected Topics in Natural Language Processing . . . . . . . . . . . . . . . . . . . . . . . . . . 167
Master Computer Science — Universität Bonn 112

MA-INF 4111 Principles of Machine Learning

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr.-Ing. Christian Bauckhage Prof. Dr.-Ing. Christian Bauckhage

Programme Mode Semester

M. Sc. Computer Science Optional 1-2.

Learning goals: technical skills

Upon successful completion of this module, students should be able to describe fundamental methods, algorithms,
and use cases of machine learning. Students acquire knowledge about supervised and unsupervised learning; based
on the knowledge and skills acquired, students should be able to
• Implement, algorithms for optimization and parameter estimation in model training and machine learning tasks.
• Adopt the fundamental methods they learned about to a wide range of problems in automated intelligent data
analysis.

Learning goals: soft skills

In the exercises, students can put their knowledge about theoretical concepts, mathematical methods, and
algorithmic approaches into practice and realize small projects involving the implementation and evaluation of
machine learning algorithms. This requires teamwork; upon successful completion of the module, students should be
able to
• draft and implement basic machine learning algorithms for various practical problem settings
• prepare and give oral presentations about their work in front of an audience

Contents
Fundamental machine learning models for classification and clustering, model training via minimization of loss
functions, fundamental optimization algorithms, model regularization, kernel methods for supervised and
unsupervised learning, probabilistic modeling and inference, dimensionality reduction and latent factor models, the
basic theory behind neural networks and neural network training; This course is intended to lay the foundation for
more advanced courses on modern deep learning and reinforcement learning.

Prerequisites
Recommended:
Linear algebra, statistics, probability theory, calculus, python programming

Remarks
At most 150 participants.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 150 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to five students. A total
of 50% of the points must be achieved. Each student must present a solution to an exercise in the exercise sessions
once.
Forms of media
• lecture slides / lecture notes are made available online
• notebooks with programming examples are made available online
Master Computer Science — Universität Bonn 113

Literature
• D.J.C MacKay: Information Theory, Inference and Learning Algorithms, Cambridge University Press, 2003
• C.M. Bishop: Pattern Recognition and Machine Learning, Springer, 2006
• S. Haykin: Neural Networks and Learning Machines, Pearson, 2008
Master Computer Science — Universität Bonn 114

MA-INF 4112 Algorithms for Data Science

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Stefan Wrobel Dr. Tamas Horvath, Prof. Dr. Stefan Wrobel

Programme Mode Semester

M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

In this module the students will learn algorithms for data science as well as implement and practice selected
algorithms from this field. The module concentrates on basic algorithms in association rule mining, graph mining,
and data streams. At the end of the module, students will be capable of analyzing formal properties of this kind of
algorithms and choosing appropriate pattern discovery and data stream algorithms.

Learning goals: soft skills

Communicative skills (oral and written presentation of solutions, discussions in teams), self-competences (ability to
accept and formulate criticism, ability to analyse, creativity in the context of an "open end" task), social skills
(effective team work and project planning).

Contents
The module is offered every year, each time concentrating on one or more specific issues, such as frequent, closed and
maximal frequent itemset mining, frequent subgraph mining algorithms for forests and for other graph classes
beyond forests, frequent items and frequency moments in data streams, and graph stream algorithms.

Prerequisites
Recommended:
Knowledge of standard notions and results from complexity theory, propositional logic, hashing, probability theory,
and calculus, all on the bachelor level, are required.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to five students. A total
of 50% of the points must be achieved. Each student must present a solution to an exercise in the exercise sessions
once.
Forms of media
lectures, exercises

Literature
• Avrim Blum, John Hopcroft, Ravindran Kannan: Foundations of Data Science. Cambridge University Press, 2020.
• Jiawei Han, Micheline Kamber, Jian Pei: Data Mining: Concepts and Techniques. Morgan Kaufmann Publishers,
2012.
• David J. Hand, Heikki Mannila and Padhraic Smyth: Principles of Data Mining. The MIT Press, 2001.
Master Computer Science — Universität Bonn 115

MA-INF 4113 Cognitive Robotics

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Sven Behnke Prof. Dr. Sven Behnke, Prof. Dr. Maren Bennewitz

Programme Mode Semester

M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

This lecture is one of two introductory lectures on Robotics of the intelligent systems track. The lecture covers
cognitive capabilities of robots, like self-localization, mapping, object perception, and action-planning in complex
environments.
This module complements MA-INF 4114 and can be taken before or after that module.

Learning goals: soft skills

Communicative skills (oral and written presentation of solutions, discussions in small teams), self competences
(ability to accept and formulate criticism, ability to analyze problems)

Contents
Probabilistic approaches to state estimation (Bayes Filters, Kalman Filter, Particle Filter), motion models, sensor
models, self-localization, mapping with known poses, simultaneous mapping and localization (SLAM), iterated
closest-point matching, path planning, place- and person recognition, object recognition.

Prerequisites
none
Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to two students. A total
of 50% of the points must be achieved.

Literature
• S. Thrun, W. Burgard and D. Fox: Probabilistic Robotics. MIT Press, 2005.
• B. Siciliano, O. Khatib (Eds.): Springer Handbook of Robotics, 2008.
• R. Szeliski: Computer Vision: Algorithms and Applications, Springer 2010.
Master Computer Science — Universität Bonn 116

MA-INF 4114 Robot Learning

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Sven Behnke Prof. Dr. Sven Behnke, Dr. Nils Goerke

Programme Mode Semester

M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

This lecture is one of two introductory lectures on Robotics of the intelligent systems track. Creating autonomous
robots that can learn to assist humans in situations of daily life is a fascinating challenge for machine learning. The
lecture covers key ingredients for a general robot learning approach to get closer towards human-like performance in
robotics, such as reinforcement learning, learning models for control, learning motor primitives, learning from
demonstrations and imitation learning, and interactive learning.
This module complements MA-INF 4113 and can be taken before or after that module.

Learning goals: soft skills

Communicative skills (oral and written presentation of solutions, discussions in small teams), self competences
(ability to accept and formulate criticism, ability to analyze problems)

Contents
Reinforcement learning, Markov decision processes, dynamic programming, Monte Carlo methods,
temporal-difference methods, function approximation, liear quadratic regulation, differential dynamic programming,
partially observable MDPs, policy gradient methods, inverse reinforcement learning, imitation learning, learning
kinematic models, perceiving and handling of objects.

Prerequisites
none
Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to two students. A total
of 50% of the points must be achieved.

Literature
• R. Sutton and A. Barto: Reinforcement Learning, MIT-Press, 1998.
• O. Sigaud and J. Peters (Eds.): From Motor Learning to Interaction Learning in Robots. Springer, 2010.
Master Computer Science — Universität Bonn 117

MA-INF 4115 Introduction to Natural Language Processing

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Lucie Flek Prof. Dr. Lucie Flek
Programme Mode Semester
M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

This class provides a technical perspective on NLP methods for building computer software that understands and
manipulates human language. Contemporary data-driven approaches are emphasized, focusing on machine learning
techniques. The covered applications vary in complexity, including for example Entity Recognition, Argument
Mining, or Emotion Analysis.

Learning goals: soft skills

Group work during programming exercises will allow students to work on real-world NLP application projects. The
final project offers you the chance to apply your newly acquired skills towards an in-depth application using different
frameworks such as PyTorch and spaCy and present it in a poster session.

Contents
Through lectures, exercises, and a final project, you will gain a thorough introduction to cutting-edge research in
NLP, from the linguistic basis of computational language methods to recent advances in deep learning and large
language models. This course provides:
• An overview of NLP goals, challenges, and applications
• Text representation (Words, sentences, paragraphs, documents), word embeddings, word2vec, BERT, word
similarity
• Machine learning / deep learning algorithms for text classification, Transformers
• Basics of neural language modeling
• Basics of computational linguistics
- Transforming words to their base forms (tokenization, stemming, lemmatization)
- Syntactic analysis (part of speech tagging, chunking, and parsing)
- Techniques for extracting meaning from text (semantic analysis), use of lexical resources in NLP
• NLP applications and projects (e.g., Sentiment Analysis, Named Entity Recognition, Question Answering,
Summarization, Fake news detection, Plagiarism detection, Abusive language detection, Opinion mining...)

Prerequisites
Recommended:
• Basics of statistics recommended.
• Basic programming knowledge in Python is of advantage.
• Basics of machine learning are of advantage.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 3 45 T / 45 S 3 S = independent study
Exercises 1 15 T / 75 S 3

Graded exams
Written exam (60 %); Project work (40 %)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work must be done individually. A total of 50% of the
points must be achieved.
Master Computer Science — Universität Bonn 118

Forms of media
• Lecture slides
• Exercise slides
• Notebooks with programming examples

Literature
• J. Eisenstein: Introduction to Natural Language Processing
• Jurafsky, Daniel, and James H. Martin. "Speech and Language Processing: An Introduction to Natural Language
Processing, Computational Linguistics, and Speech Recognition."
• S. Bird, E. Klein, E. Loper; Natural Language Processing with Python
Master Computer Science — Universität Bonn 119

MA-INF 4116 Seminar AI Safety

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Lucie Flek Prof. Dr. Lucie Flek, Dr. Florian Mai

Programme Mode Semester

M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

The seminar aims to introduce students to cutting edge research problems around the safety, alignment, and
ethicality of advanced AI systems. Students will acquire knowledge of common safety-related concerns, develop skills
in assessing the safety of AI systems, and how to improve the robustness, alignment, transparency and safety of
state-of-the-art AI models such as large language models and agentic systems. Students will learn to understand,
analyze, and communicate the motivation, methodology, and results of technical papers from the AI safety field.
Through the writing of an essay on one of the course topics, they will learn to conduct a literature review, identify
gaps in the present literature, and devise a plan to advance the scientific frontier.

Learning goals: soft skills

Contents
As AI systems such as Large Language Models become increasingly capable and start to be used in high-stakes
scenarios, ensuring that they act safely is gaining importance. The research field of AI safety studies methods to
align the behavior and values of AI systems with the user and broader society in a robust, scalable, and interpretable
way. It studies methods that monitor AI systems, detect misbehavior, and correct or prevent it if necessary. The aim
of this course is to explore cutting-edge research, insights, and trends in the field of AI safety.
A typical lecture will consist of 2-3 student presentations that focus on a research article and the broad context of its
topic. Following each presentation, we discuss the work with a focus on assessing relevant safety issues and potential
approaches for safer design and engineering.

Prerequisites
Recommended:
One of the following courses is recommended:
• MA-INF 4115 - Introduction to Natural Language Processing
• MA-INF 4332 - Seminar Large Language Models
• MA-INF 4335 - Lab AI Alignment

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, final essay

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
• Amodei, Dario: Concrete Problems in AI Safety
• Hendrycks, Dan: What would jiminy cricket do? towards agents that behave morally
• Ouyang, Long et al.: Training language models to follow instructions with human feedback
• Burns, Colins et al.: Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision
• Betley, Jan: Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs
• Bricken, Trenton et al.: Towards monosemanticity: Decomposing language models with dictionary learning
Master Computer Science — Universität Bonn 120

MA-INF 4117 Mining Media Data I

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Rafet Sifa Prof. Dr. Rafet Sifa, Dr. Lorenz Sparrenberg

Programme Mode Semester

M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

By the end of this course, students will be able to:
• Analyze and extract meaningful relationships from large-scale media datasets using advanced data mining
techniques.
• Develop and implement predictive and descriptive models for applications such as recommender systems, trend
analysis, and outlier detection.
• Apply mathematical optimization methods to create interpretable and efficient machine learning models.
• Integrate theoretical concepts with practical tools to address challenges in digital forensics, behavioral profiling,
and marketing strategy design.

Learning goals: soft skills

Contents
This course provides a comprehensive exploration of advanced data mining techniques tailored for media data
analysis. Students will delve into methods like affinity mining, latent pattern mining, neural networks, and
archetypal analysis to uncover insights in behavioral profiling, recommender systems, and outlier detection.
Emphasis is placed on theoretical understanding and practical application through mathematical optimization,
interpretable models, and real-world case studies, enabling participants to harness data for impactful digital
marketing, fraud detection, and content personalization.

Prerequisites
Recommended:
Basic knowledge of data science, machine learning, and pattern recognition; programming skills; linear algebra.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 1 15 T / 30 S 1.5

Graded exams
Written exam (120 minutes) or oral exam.

Ungraded coursework (required for admission to the exam)

(i) The completion of regularly provided exercise sheets. The work can be done in groups of up to five students. A
total of 50% of the points must be achieved. (ii) The completion of a programming project. The work is done in
groups of up to five students. The results of the programming project must be presented in class.

Literature
• Rafet Sifa (2019). Matrix and Tensor Factorization for Profiling Player Behaviour. Independently pressed.
• Jiawei Han, Jian Pei, Micheline Kamber (2012). Data Mining: Concepts and Techniques. 3rd Ed., Elsevier Inc.
• Christopher Manning, Prabhakar Raghavan, Hinrich Schütze (2008). Introduction to Information Retrieval.
Cambridge University Press.
• Michael Negnevitsky (2019), Artificial Intelligence: A Guide to Intelligent Systems. 3rd Ed., Pearson.
• Ian Goodfellow, Yoshua Bengio, Aaron Courville (2016). Deep Learning. MIT Press..
• Lecture notes of the instructors.
Master Computer Science — Universität Bonn 121

MA-INF 4201 Artificial Life

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Sven Behnke Dr. Nils Goerke
Programme Mode Semester
M. Sc. Computer Science Optional 1-3.

Learning goals: technical skills

Detailed understanding of the most important approaches and principles of artificial life. Knowledge and
understanding of the current state of research in the field of artificial life. The students can judge and explain if an
Artificial Life approach is feasible for a given class of problems. They can estimate the necessary effort to implement
and shape the Artificial Life paradigm w.r.t. the task, and can give an educated estimation of the possible outcome
and forseeable limitations of the approach. They can implement the basic fundamental Artificial Life paradigms.

Learning goals: soft skills

Capability to identify the state of the art in artificial life, and to present and defend the found solutions within the
exercises in front of a group of students. Critical discussion of the results of the homework.

Contents
Foundations of artificial life, cellular automata, Conway’s “Game of Life”; mechanisms for structural development;
foundations of nonlinear dynamical systems, Lindenmeyer-systems, evolutionary methods and genetic algorithms,
reinforcement learning, artificial immune systems, adaptive behaviour, self-organising criticality, multi-agent systems,
and swarm intelligence, particle swarm optimization.

Prerequisites
Recommended:
Basic knowledge of linear algebra, analysis, logic, automata, and complexity analysis of deterministic and
randomised algorithms.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (100 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to four students. A total
of 50% of the points must be achieved. Each student must present a solution to an exercise in the exercise sessions
twice.
Forms of media
Pencil and paper work, explain solutions in front of the exercise group, implementation of small programs, use of
simple simulation tools.

Literature
• Christoph Adami: Introduction to Artificial Life, The Electronic Library of Science, TELOS, Springer-Verlag
• Eric Bonabeau, Marco Dorigo, Guy Theraulaz: Swarm Intelligence: From Natural to Artificial Systems, Oxford
University Press, Santa Fe Institute Studies in the Science of Complexity.
• Andrzej Osyczka: Evolutionary Algorithms for Single and Multicriteria Design Optimization, Studies in Fuzzyness
and Soft Computing, Physica-Verlag, A Springer-Verlag Company, Heidelberg
Master Computer Science — Universität Bonn 122

MA-INF 4204 Technical Neural Nets

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Dr. Nils Goerke Dr. Nils Goerke
Programme Mode Semester
M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

Detailed knowledge of the most important neural network approaches and learning algorithms and its fields of
application. Knowledge and understanding of technical neural networks as Non-Von Neumann computer
architectures similar to concepts of brain functions at different stages of development. The students can judge and
explain if a neural network approach is feasible for a given class of problems. They can estimate the necessary effort
to implement and shape the neural approach for a given task and can give an educated estimation of the possible
outcome and foseeable limitations of that approach. They can implement the basic neural network approaches and
neural learning paradigms.

Learning goals: soft skills

The students will be capable to propose several paradigms from neural networks that are capable to solve a given
task. They can discuss the pro and cons with respect to efficency and risk. The will be capable to plan and
implement a small project with state of the art neural network solutions. Capability to identify the state of the art
in neural network research. Capability to present and defend the found solutions within the exercises in front of a
group of students. Critical discussion of the results of the homework.

Contents
Multi-layer perceptron, radial-basis function nets, Hopfield nets, self organizing maps (Kohonen), adaptive resonance
theory, learning vector quantization, recurrent networks, back-propagation of error, reinforcement learning,
Q-learning, support vector machines, pulse processing neural networks. Exemplary applications of neural nets:
function approximation, prediction, quality control, image processing, speech processing, action planning, control of
technical processes and robots. Implementation of neural networks in hardware and software: tools, simulators,
analog and digital neural hardware.

Prerequisites
Recommended:
Basic knowledge of linear algebra, analysis, logic, automata, complexity analysis of deterministic and randomised
algorithms, and practical and theoretical foundations of machine learning.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (100 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to four students. A total
of 50% of the points must be achieved. Each student must present a solution to an exercise in the exercise sessions
twice.
Forms of media
Pencil and paper work, explaining solutions in front of the exercise group, implementation of small programs, use of
simple simulation tools
Master Computer Science — Universität Bonn 123

Literature
• Christopher M. Bishop: Neural Networks for Pattern Recognition, Oxford University Press, ISBN-10: 0198538642,
ISBN-13: 978-0198538646
• Ian T. Nabney: NETLAB. Algorithms for Pattern Recognition, Springer, ISBN-10: 1852334401, ISBN-13:
978-1852334406
• David Kriesel: A brief Introduction on Neural Networks, [Link]
• David Kriesel: Ein kleiner Überblick über Neuronale Netze, [Link]
• Simon Haykin: Neural Networks, and Learning Machines, 3rd Edition, Prentice Hall International Editions.
Master Computer Science — Universität Bonn 124

MA-INF 4208 Seminar Vision Systems

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Sven Behnke Prof. Dr. Sven Behnke, Dr. Nils Goerke

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

• Knowledge in advanced topics in the area of technical vision systems, such as image segmentation, feature
extraction, and object recognition.
• Ability to understand new research results presented in original scientific papers and to present them in a research
talk as well as in a seminar report.

Learning goals: soft skills

Self-competences (time management, literature search, self-study), communication skills (preparation and clear
didactic presentation of research talk, scientific discussion, structured writing of seminar report), social skills (ability
to formulate and accept criticism, critical examination of research results).

Contents
Current research papers from conferences and journals in the field of vision systems covering fundamental techniques
and applications.

Prerequisites
Recommended:
At least one of the following:
• MA-INF 2201 - Computer Vision
• MA-INF 4111 – Principles of Machine Learning
• MA-INF 4204 – Technical Neural Nets
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
• R. Szeliski: Computer Vision: Algorithms and Applications, Springer 2010.
• C. M. Bishop: Pattern Recognition and Machine Learning, Springer 2006.
• D. A. Forsyth and J. Ponce: Computer Vision: A Modern Approach, Prentice Hall, 2003.
Master Computer Science — Universität Bonn 125

MA-INF 4209 Seminar Principles of Data Mining and Learning Algorithms

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Stefan Wrobel Prof. Dr. Stefan Wrobel, PD Dr. Michael Mock, Dr. Florian Seiffarth,
Dr. Tamas Horvath
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Enhanced and in-depth knowledge in specialized topics in the area of machine learning and data mining.

Learning goals: soft skills

Acquire the competence to independently search for and study state-of-the-art scientific literature in depth, read
critically, identify the most relevant content, and assess research results in the context of the corresponding research
area; to discuss research results with a knowledgeable scientific audience; to present prior work by others in writing
and in presentations with visual media in a way that adheres to academic standards, that is well-structured and
didactically effective, and that motivates the audience to participate; to formulate and accept criticism; to manage
one’s time with relatively open assignments and long-ranging deadlines.

Contents
Theoretical, statistical and algorithmical principles of data mining and learning algorithms. Search and optimization
algorithms. Specialized learning algorithms from the frontier of research. Fundamental results from neighbouring
areas.
Prerequisites
Recommended:
Knowledge of basic notions and algorithms from machine learning and data mining. It is recommend to first take at
least one of the following modules:
• MA-INF 4111 – Principles of Machine Learning
• MA-INF 4112 – Algorithms for Data Science

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Forms of media
Scientific papers and websites, interactive presentations.

Literature
The relevant literature will be announced towards the end of the previous semester.
Master Computer Science — Universität Bonn 126

MA-INF 4211 Seminar Cognitive Robotics

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Sven Behnke Prof. Dr. Sven Behnke, Dr. Raphael Memmesheimer

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Knowledge in advanced topics in the area of cognitive robotics, such as robot perception, action planning, and robot
learning.
Ability to understand new research results presented in original scientific papers and to present them in a research
talk as well as in a seminar report.

Learning goals: soft skills

Self-competences (time management, literature search, self-study), communication skills (preparation and clear
didactic presentation of research talk, scientific discussion, structured writing of seminar report), social skills (ability
to formulate and accept criticism, critical examination of research results).

Contents
Current research papers from conferences and journals in the field of cognitive robotics covering fundamental
techniques and applications.

Prerequisites
Recommended:
At least 1 of the following:
• MA-INF 4113 – Cognitive Robotics
• MA-INF 4114 – Robot Learning

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
• S. Thrun, W. Burgard and D. Fox: Probabilistic Robotics. MIT Press, 2005.
• B. Siciliano, O. Khatib (Eds.): Springer Handbook of Robotics, 2008.
• Selected papers.
Master Computer Science — Universität Bonn 127

MA-INF 4213 Seminar Humanoid Robots

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Maren Bennewitz Prof. Dr. Maren Bennewitz
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Enhanced and in-depth knowledge in specialized topics in the area of humanoid robotics, such as perception, state
estimation, navigation, manipulation, and motion planning.

Learning goals: soft skills

Self-competences (time management, literature search, self-study), communication skills (preparation of the talk,
clear didactic presentation of techniques and experimental results, scientific discussion, structured writing of
summary), social skills (ability to formulate and accept criticism, critical examination of algorithms and
experimental results).

Contents
Current research papers from conferences and journals in the field of humanoid robotics covering fundamental
techniques and applications.

Prerequisites
Recommended:
At least 1 of the following:
• MA-INF 4215 – Humanoid Robotics
• MA-INF 4113 – Cognitive Robotics

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
• S. Thrun, W. Burgard and D. Fox: Probabilistic Robotics. MIT Press
• B. Siciliano, O. Khatib (Eds.): Springer Handbook of Robotics
• K. Harada, E. Yoshida, K. Yokoi (Eds.), Motion Planning for Humanoid Robots, Springer
• Selected papers.
Master Computer Science — Universität Bonn 128

MA-INF 4214 Lab Humanoid Robots

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Maren Bennewitz Prof. Dr. Maren Bennewitz
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Design and implementation of perception, state estimation, navigation, manipulation, and motion planning
techniques for humanoid robots.

Learning goals: soft skills

Ability to analyze problems theoretically and to find creative and practical solutions; to examine one’s solutions and
results critically; to prepare readable documentation of software and research results; to present, defend and discuss
design decisions and results in the team/group and to other students clearly and in accordance with academic
standards; to collaborate constructively with others in small teams over a longer period of time;

Contents
Robot middleware, perception, state estimation, navigation, manipulation, and motion planning for humanoid
robots.
Prerequisites
Recommended:
At least 1 of the following:
• MA-INF 4215 – Humanoid Robotics
• MA-INF 4113 – Cognitive Robotics

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
• S. Thrun, W. Burgard and D. Fox: Probabilistic Robotics. MIT Press
• B. Siciliano, O. Khatib (Eds.): Springer Handbook of Robotics
• K. Harada, E. Yoshida, K. Yokoi (Eds.), Motion Planning for Humanoid Robots, Springer
• Selected papers.
Master Computer Science — Universität Bonn 129

MA-INF 4215 Humanoid Robotics

Workload Credit points Duration Frequency

180 h 6 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Maren Bennewitz Prof. Dr. Maren Bennewitz
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

This lecture covers techniques for humanoid robots such as perception, navigation, and motion planning. After the
lecture, the students will be able to understand and implement techniques that enable humanoid robots to
autonomously navigate in human environments as well as perceive, represent, and manipulate objects.

Learning goals: soft skills

Communicative skills (oral and written presentation of solutions, discussions in small teams), ability to analyze
problems.

Contents
Sensing and perception, environment representations, active perception, inverse kinematics, motion planning,
grasping, balance control, walking, and footstep planning.

Prerequisites
Recommended:
MA-INF 4113 – Cognitive Robotics

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (90 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to three students. A
total of 50% of the points must be achieved.

Literature
• S. Thrun, W. Burgard and D. Fox: Probabilistic Robotics. MIT Press, 2005.
• B. Siciliano, O. Khatib (Eds.): Springer Handbook of Robotics
• K. Harada, E. Yoshida, K. Yokoi (Eds.), Motion Planning for Humanoid Robots, Springer
• Selected research papers.
Master Computer Science — Universität Bonn 130

MA-INF 4216 Biomedical Data Science and AI

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Dr. Holger Fröhlich Dr. Holger Fröhlich

Programme Mode Semester

M. Sc. Computer Science Optional 3.

Learning goals: technical skills

- understanding and knowledge of fundamental data mining and machine learning methods
- understanding of their application in bioinformatics

Learning goals: soft skills

- communication: oral and written presentation of solutions to exercises
- self-competences: ability to analyze application problems and to formulate possible solutions
- practical skills: ability to practically implement solutions
- social skills: working in a small team with other students

Contents
This lecture gives a broad overview about frequently used statistical techniques as well as data mining and machine
learning algorithms in bioinformatics. The use of the respective methods to solve problems in bioinformatics is
explained. The goal is to understand the explained methods, being able to apply them correctly and partially
implement them. More detailed, the following topics are covered in the context of their application in bioinformatics:
- Short introduction to Bioinformatics and Biomedicine
- Statistical Basics: Probability distributions and Bayesian inference, statistical hypothesis testing, linear models,
logistic regression, Principal Component Analysis
- Clustering
- Hidden Markov Models
- Supervised Machine Learning
- Elastic Net
Prerequisites
none
Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam
Ungraded coursework (required for admission to the exam)
(i) The completion of regularly provided exercise sheets. The work can be done in groups of up to three students. A
total of 50% of the points must be achieved. Each student must present a solution to an exercise in the exercise
sessions once. (ii) Participation in an achievement test. On the test, at least 50% of the points much be achieved.

Literature
T. Hastie, R. Tibshirani, J. Friedman, The Elements of Statistical Learning, Springer, 2008
[Link], P. Watters, Statistics in a Nutshell, O’Reilly, 2008
N. Jones, P. Pevzner, An Introduction to Bioinformatics Algorithms, MIT Press, 2004
Master Computer Science — Universität Bonn 131

MA-INF 4217 Seminar Machine Learning Methods in the Life Sciences

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

Dr. Holger Fröhlich Dr. Holger Fröhlich

Programme Mode Semester

M. Sc. Computer Science Optional 4.

Learning goals: technical skills

- understanding and knowledge of machine learning methods and their application in modern life sciences, e.g.
biomedicine
Learning goals: soft skills
- communication: oral scientific presentation of a defined topic
- self-competences: ability to identify relevant literature for a given topic; ability to read, understand and analyze
scientific publications
- social skills: ability to discuss a scientific topic with other students and the staff

Contents
Machine learning techniques play a crucial role in modern life sciences, including biomedicine. The goal of this
seminar is to discuss a variety of machine learning techniques in the context of their application to solve real-world
problems in biomedicine.
Topics will be selected from the following areas:
- Ensemble learning
- Survival and disease progression models
- Bayesian Networks
- Stochastic processes, e.g. Gaussian Proceses, Dirichlet Process Mixture Models
- MCMC methods
- Deep learning methods, e.g. DNNs, CNNs, Deep Belief Networks
- feature selection and non-linear embedding methods
- multi-modal data fusion techniques
Attendees will be asked to perform research about their topic in a self-responsible manner.

Prerequisites
Recommended:
MA-INF 4216 – Data Mining and Machine Learning Methods in Bioinformatics

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Forms of media
powerpoint

Literature
selected journal and conference papers
Master Computer Science — Universität Bonn 132

MA-INF 4226 Lab Parallel Computing for Mobile Robotics

Workload Credit points Duration Frequency

270 h 9 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Maren Bennewitz Prof. Dr. Maren Bennewitz
Programme Mode Semester
M. Sc. Computer Science Optional 2.

Learning goals: technical skills

Students will make practical experience with the design and implementation of parallelized algorithms in the context
of motion planning and navigation.

Learning goals: soft skills

Ability to properly present and defend design decisions, to prepare readable documentation of software; skills in
constructively collaborating with others in small teams over a longer period of time; ability to classify ones own
results into the state-of-the-art of the resp. area

Contents
Parallel programming on the GPU, CUDA, shortest path planning, collision checking, visibility graph, A* algorithm

Prerequisites
Recommended:
C++, Linux.
Since the exercises revolve around path planning, one of those courses might be helpful:
MA-INF 4203: Autonomous Mobile Systems
MA-INF 4113: Cognitive Robotics
MA-INF 4310: Lab Mobile Robots
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 133

MA-INF 4228 Foundations of Data Science

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Lucie Flek Prof. Dr. Lucie Flek
Programme Mode Semester
M. Sc. Computer Science Optional 1-3.

Learning goals: technical skills

Knowledge: Peculiarities of high dimensional spaces in geometry and probabilities. Singular vector decomposition.
Basics in machine learning and clustering.
Skills: Understanding of mathematical tools.
Competences: Application to data science problems and ability to assess similar methods.

Learning goals: soft skills

Oral presentation (in tutorial groups), written presentation (of exercise solutions), team collaboration in solving
homework problems, critical assessmen

Contents
Data science aims at making sense of big data. To that end, various tools have to be understood for helping in
analyzing the arising structures.
Often data comes as a collection of vectors with a large number of components. To understand their common
structure is the first main objective of understanding the data. The geometry and the linear algebra behind them
becomes relevant and enlightning. Yet, the intuition from low-dimensional space turns out to be often misleading.
We need to be aware of the particular properties of high-dimensional spaces when working with such data. Fruitful
methods for the analysis include singular vector decomposition from linear algebra and supervised and unsupervised
machine learning.

Prerequisites
Recommended:
Basic skills in linear algebra and stochastics.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 4 60 T / 105 S 5.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. A total of 50% of the points must be achieved. Each student
must present a solution to an exercise in the exercise sessions twice.

Literature
Avrim Blum, John Hopcroft, and Ravindran Kannan (2018+). Foundations of Data Science.
Master Computer Science — Universität Bonn 134

MA-INF 4230 Advanced Methods of Information Retrieval

Workload Credit points Duration Frequency

180 h 6 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Elena Demidova Prof. Dr. Elena Demidova
Programme Mode Semester
M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

This module introduces the students to the advanced methods, data structures, and algorithms of information
retrieval for structured and semi-structured data (including, for example, knowledge graphs, relational data, and
tabular data).
At the end of the module, the students will be capable of choosing appropriate data structures and retrieval
algorithms for specific applications and correctly apply relevant statistical and machine learning-based information
retrieval procedures.

Learning goals: soft skills

Communication skills: oral and written presentation and discussion of solutions.
Self-competences: ability to analyse and solve problems.

Contents
The module topics include data structures, ranking methods, and efficient algorithms that enable end-users to
effectively obtain the most relevant search results from structured, heterogeneous, and distributed data sources.
Furthermore, we will study the corresponding evaluation techniques as well as novel applications.

Prerequisites
Recommended:
Basic knowledge of data science and machine learning; programming skills. Recommended reading:
• Sarah Boslaugh. Statistics in a Nutshell. A Desktop Quick Reference, O’Reilly Media, Inc., 2nd Edition, (2012).
• Ethem Alpaydin. Machine Learning. The MIT Press (2021).

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to three, four or five
students, depending on the total number of students taking the course. A total of 50% of the points must be
achieved. For 80% of the exercise sheets, 40% of the points must be achieved for each sheet. Each student must
present a solution to an exercise in the exercise sessions once.
Master Computer Science — Universität Bonn 135

Literature
Selected chapters from:
• Christopher D. Manning, Prabhakar Raghavan and Hinrich Schütze, Introduction to Information Retrieval,
Cambridge University Press. 2008.
• Bhaskar Mitra and Nick Craswell (2018), "An Introduction to Neural Information Retrieval ", Foundations and
Trendső in Information Retrieval: Vol. 13: No. 1, pp 1-126.
• Ridho Reinanda, Edgar Meij and Maarten de Rijke (2020), "Knowledge Graphs: An Information Retrieval
Perspective", Foundations and Trendső in Information Retrieval: Vol. 14: No. 4, pp 289-444.
• Jeffrey Xu Yu, Lu Qin, Lijun Chang. Keyword Search in Databases. Synthesis Lectures on Data Management.
Morgan & Claypool Publishers. 2009.
Further references to relevant material will be provided during the lecture.
Master Computer Science — Universität Bonn 136

MA-INF 4231 Seminar Advanced Topics in Information Retrieval

Workload Credit points Duration Frequency

120 h 4 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Elena Demidova Prof. Dr. Elena Demidova
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Enhanced and in-depth knowledge in specialized topics in the area of information retrieval, including understanding
of information retrieval process, specialized data representation methods, advanced retrieval methods, evaluation
techniques, and domain-specific applications.

Learning goals: soft skills

Acquire the competence to independently search for and study state-of-the-art scientific literature in depth, read
critically, identify the most relevant content, and assess research results in the context of the corresponding research
area; to discuss research results with a knowledgeable scientific audience; to present prior work by others in writing
and in presentations with visual media in a way that adheres to academic standards, that is well-structured and
didactically effective, and that motivates the audience to participate; to formulate and accept criticism; to manage
one’s time with relatively open assignments and long-ranging deadlines.

Contents
Statistical and machine learning-based information retrieval methods, including typical steps of the information
retrieval process: data collection, feature extraction, indexing, retrieval, ranking, and evaluation. Specialized data
representation and retrieval methods for selected data types and applications in specific domains.

Prerequisites
Recommended:
MA-INF 4230 - Advanced Methods of Information Retrieval.
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
Selected chapters from:
• Christopher D. Manning, Prabhakar Raghavan and Hinrich Schütze, Introduction to Information Retrieval,
Cambridge University Press. 2008.
• Bhaskar Mitra and Nick Craswell (2018), "An Introduction to Neural Information Retrieval ", Foundations and
Trendső in Information Retrieval: Vol. 13: No. 1, pp 1-126.
Further relevant literature will be announced at the beginning of the seminar.
Master Computer Science — Universität Bonn 137

MA-INF 4232 Lab Information Retrieval in Practice

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Elena Demidova Dr. Rajjat Dadwal

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

This module concentrates on practical experience in information retrieval. Participants acquire basic knowledge and
practical experience in designing and implementing information retrieval systems for specific data types and
applications.

Learning goals: soft skills

Communication skills: the ability to work in teams.
Self-competences: the ability to analyse problems and find practical solutions. Time management, creativity,
presentation of results.

Contents
Practical application of information retrieval methods to solve retrieval problems on real-world data and evaluate
proposed solutions.

Prerequisites
Recommended:
MA-INF 4230 - Advanced Methods of Information Retrieval
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
Selected chapters from:
• Christopher D. Manning, Prabhakar Raghavan and Hinrich Schütze, Introduction to Information Retrieval,
Cambridge University Press. 2008.
• Bhaskar Mitra and Nick Craswell (2018), "An Introduction to Neural Information Retrieval ", Foundations and
Trendső in Information Retrieval: Vol. 13: No. 1, pp 1-126.
Further references to relevant material will be provided during the lab.
Master Computer Science — Universität Bonn 138

MA-INF 4235 Reinforcement Learning

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every 2 years

Module coordinator Lecturer(s)

Prof. Dr.-Ing. Christian Bauckhage Prof. Dr.-Ing. Christian Bauckhage

Programme Mode Semester

M. Sc. Computer Science Optional 2-3.

Learning goals: technical skills

Upon successful completion of this module, students should be able to describe fundamental methods, algorithms,
and use cases of reinforcement learning. Students acquire knowledge about underlying mathematical models and
corresponding algorithms; based on the knowledge and skills acquired, students should be able to:
• implement algorithms for reinforcement learning problems;
• adopt the fundamental methods they learned about to a wide
range of problems in policy optimization.

Learning goals: soft skills

In the exercises, students can put their knowledge about theoretical concepts, mathematical methods, and
algorithmic approaches into practice and realize small projects involving the implementation and evaluation of
search- and policy learning algorithms. This requires teamwork; upon successful completion of the module, students
should be
able to:
• draft and implement basic reward functions and policy learning algorithms for various practical problem settings;
• prepare and give oral presentations about their work in front of an audience.

Contents
State space models, tree search algorithms, Monte Carlo tree search,
Markov chain models, Markov decision processes, value functions,
reward functions, Bellman equations, policy learning, TD learning Q
learning, deep Q learning

Prerequisites
Required:
Linear algebra, statistics, probability theory, python programming

Remarks
At most 150 participants.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam
Ungraded coursework (required for admission to the exam)
The completion of regularly provided exercise sheets. The work can be done in groups of up to six students. Each
student must present a solution to an exercise in the exercise sessions once. There is one mid-term test where a total
of 50% of the credits must be achieved.

Forms of media
• lecture slides / lecture notes are made available online
• notebooks with programming examples are made available online
Master Computer Science — Universität Bonn 139

Literature
R.S. Sutton and A.G. Barto: Reinforcement Learning, 2nd ed., MIT Press,
2018
Master Computer Science — Universität Bonn 140

MA-INF 4237 Lab Natural Language Processing

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Lucie Flek Prof. Dr. Lucie Flek
Programme Mode Semester
M. Sc. Computer Science Optional 2-3.

Learning goals: technical skills

The Natural Language Processing (NLP) Lab course provides students with a detailed look at the recent
advancements in NLP, covering various aspects such as large language models (LLMs), conversational systems, and
computational social science. The course emphasizes a practical approach and offers you the opportunity to gain
hands-on experience in developing NLP-based systems, allowing you to deepen your understanding of NLP
technologies and apply theoretical knowledge to real-world scenarios.

Learning goals: soft skills

Through tutorials and a final project, you will gain practical skills in NLP techniques and have this chance to apply
this knowledge to a various interesting project. Students will collaborate in small teams (a group of two students) and
implement NLP applications over the course of the term. Each team is advised by one researcher of the CAISA Lab.

Contents
The course emphasizes a practical approach and offers you the opportunity to gain hands-on experience in
developing NLP-based systems, allowing you to deepen your understanding of NLP technologies and apply
theoretical knowledge to real-world scenarios.

Prerequisites
Required:
MA-INF 4115: Introduction to Natural Language Processing
Recommended:
• Basic programming knowledge in Python and Machine Learning
• Basics of Machine Learning
• Basic knowledge of Python Libraries for ML (NumPy, Scikit-Learn, Pandas)
• Basics of Probability, Linear Algebra and Statistics

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 141

MA-INF 4240 Lab Hybrid Learning and Applications

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Rafet Sifa Prof. Dr. Rafet Sifa, Dr. Lorenz Sparrenberg

Programme Mode Semester

M. Sc. Computer Science Optional 2-4.

Learning goals: technical skills

• Studying a self-selected research topic
• Reproducing important results
• Elaborating findings based on own research
• Applying theoretical knowledge to real-world problems
• Familiarity with external research work

Learning goals: soft skills

• Own idea generation
• Project completion within self-defined scope and timeline
• Adapting relevant aspects to own projects
• Communication skills through structured presentations

Contents
This lab offers a comprehensive introduction to using hybrid learning, merging machine learning and deep learning
techniques to address complex problems. By integrating foundation models with downstream tasks using various
machine learning methods, students explore a range of fascinating applications. They are encouraged to select and
research their own project topics, gaining hands-on experience in data preprocessing, model building, evaluation, and
optimization. This course is designed to equip students with practical skills to design and implement effective hybrid
learning solutions.
Schedule:
1. organization meeting
2. presentation of the research idea and its application (1 week later)
3. midterm presentation of results
4. final presentation
5. Student paper

Prerequisites
Required:
• Independent work required
Recommended:
• A basic understanding of machine learning is helpful
• Students should bring their own ideas.

Remarks
Due to the limit of 10 participants, students must send their participation request and a few sentences about their
research idea to amllab@[Link] before the first appointment. Places will be allocated according to the date
of receipt and the quality of the idea submitted.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report
Master Computer Science — Universität Bonn 142

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
• Topic dependent and specified or researched by the student
• Lecture notes of the instructor (Advanced methods for text mining by Prof. Dr. Rafet Sifa, SS24)
Master Computer Science — Universität Bonn 143

MA-INF 4241 Lab Cognitive Modelling of Biological Agents

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Dr. Dominik Bach Prof. Dr. Dr. Dominik Bach
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

• Cognitive modelling workflow in computational neuroscience.
• Analysis of real-life cognitive tasks.
• Reasoning about different problem solutions.
• Understanding constraints of biological systems.

Learning goals: soft skills

Ability to analyze problems theoretically and to find creative and practical solutions; to examine one’s solutions and
results critically; to classify one’s own results into the state-of-the-art of the respective area; to prepare readable
documentation of software and research results; to present, defend and discuss design decisions and results in the
team/group and to other students clearly and in accordance with academic standards; to collaborate constructively
with others in small teams over a longer period of time; to aim at long-range goals under limited resources; to work
under pressure.

Contents
The goal of cognitive modelling in computational neuroscience is to reverse-engineer how a real neural system solves
a given cognitive task, often using reinforcement learning theory as a starting point. This lab covers the entire
cognitive modelling workflow as used in computational neuroscience. Students will address an interesting cognitive
problem by (a) developing rational solutions drawing on reinforcement learning, or descriptive solutions drawing on
cognitive science and mathematical psychology, (b) derive behavioural signatures of this solution by mathematical
analysis or computational simulation, (c) design efficient experiments to disambiguate these solutions from real
behaviour, and (d) potentially analyse existing data sets. The course emphasises a practical, application-focused
approach. Students collaborate in teams of 2, each supervised by a CAIAN researcher.

Prerequisites
Recommended:
One out of:
• MA-INF 4113 Cognitive Robotics
• MA-INF 4114 Robot Learning
• MA-INF 4215 Humanoid Robotics
• MA-INF 4235 Reinforcement Learning

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 144

MA-INF 4242 Self-supervised Learning

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

JProf. Dr. Hermann Blum JProf. Dr. Hermann Blum
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Students can list and explain different fundamental unsupervised, self-supervised machine learning models and can
justify a choice of model for a given problem. Students can list and explain different techniques of self-supervision
and semi-supervision. For a given data source and task, they can evaluate if a technique is applicable and implement
it based on established software frameworks.
Learning goals: soft skills
Communicative skills (discussions in small teams), self competences (ability to accept and formulate criticism, ability
to analyze problems).

Contents
The course covers a broad range of methods to learn from unlabelled data with a focus on deep neural networks and
hybrid methods.
Models: Nearest Neighbors, Clustering, Autoencoders, (pretrained) Neural Networks, Generative Models
Self-Supervision Techniques: Denoising, Masking, Pre-text Tasks, Distillation, Cycleconsistency, Contrastive
Learning, Pseudo-Labelling, System-level Self-Supervision.

Prerequisites
Recommended:
It is recommended that the students have:
• knowledge of foundations of artifical intelligence and deep learning.
It is recommended to first complete the following module:
• MA-INF 4111 Principles of Machine Learning.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 min)

Ungraded coursework (required for admission to the exam)

none
Literature
• scikit-learn User Guide.
• I. Goodfellow, Y. Bengio and A. Courville: Deep Learning. MIT Press, 2016.
• Balestriero et al.: A Cookbook of Self-Supervised Learning, 2023.
Master Computer Science — Universität Bonn 145

MA-INF 4243 Mining Media Data II

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Rafet Sifa Prof. Dr. Rafet Sifa, Dr. Lorenz Sparrenberg

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

By the end of the course, students will be able to:
• Understand and implement advanced data mining techniques for predictive and prescriptive analytics.
• Employ large language models and transformer-based architectures for tasks like text analysis, classification, and
summarization.
• Apply knowledge distillation techniques to optimize and deploy machine learning models in resource-constrained
environments.
• Analyze media data effectively to derive insights and support decision-making in real-world applications, including
digital marketing and fraud detection.
• Address challenges in media analytics, such as ethical considerations, model interpretability, and efficient resource
use.
Learning goals: soft skills

Contents
This course explores advanced techniques in data mining, emphasizing predictive and prescriptive methods applied
to media data. Students will learn to analyze large and complex datasets using state-of-the-art machine learning
methodologies, including behavioral prediction, knowledge distillation, and large language models (LLMs). The
curriculum includes foundational concepts, text representation learning, transformer architectures, and practical
applications in media analytics, such as recommendation systems and information extraction. The course combines
theoretical instruction with hands-on exercises to develop both technical and analytical skills relevant to industry
and research.
Prerequisites
Recommended:
Basic knowledge of data science, machine learning, and pattern recognition; programming skills; linear algebra.
Completion of MA-INF 4117 - Mining Media Data I is recommended but not mandatory.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 1 15 T / 30 S 1.5

Graded exams
Written exam (120 min) or oral exam.

Ungraded coursework (required for admission to the exam)

(i) The completion of regularly provided exercise sheets. The work can be done in groups of up to five students. A
total of 50% of the points must be achieved. (ii) The completion of a programming project. The work is done in
groups of up to five students. The results of the programming project must be presented in class.
Master Computer Science — Universität Bonn 146

MA-INF 4244 Lab Deep Learning for the Physical Sciences

Workload Credit points Duration Frequency

270 h 9 CP 1 semester at least every year

Module coordinator Lecturer(s)

JProf. Dr. Zorah Lähner JProf. Dr. Zorah Lähner, Dr. Alexander Rüttgers

Programme Mode Semester

M. Sc. Computer Science Optional 2-3.

Learning goals: technical skills

Ability to apply theoretical knowledge in practical applications (mostly from physical sciences), to extract relevant
information from research publications, to read research publications from an interdisciplinary point of view, to
implement and visualize experimental setups.

Learning goals: soft skills

Ability to analyze problems theoretically and to find creative and practical solutions; to examine one’s solutions and
results critically; to classify one’s own results into the state-of-the-art of the respective area; to prepare readable
documentation of software and research results; to present, defend and discuss design decisions and results in the
team/group and to other students clearly and in accordance with academic standards; to collaborate constructively
with others in small teams over a longer period of time.

Contents
This lab gives students the opportunity to apply their knowledge on methods of deep learning on real applications
drawn from the physical sciences, e.g. astro physics or chemistry. Students will learn about a specific problem from a
different field and potential approaches to tackle it, while gaining hands-on experience in data (pre-)processing,
model building, evaluation, and optimization, and being encouraged to proposes their own solutions.

Prerequisites
Recommended:
Previous experience with deep learning algorithms is vital, e.g. by
• MA-INF 4333 - Geometric Deep Learning
• BA-INF 153 - Einführung in Deep Learning für Visual Computing

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 147

MA-INF 4245 Lab AI for Scientific Discovery

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Vahid Babaei Prof. Dr. Vahid Babaei
Programme Mode Semester
M. Sc. Computer Science Optional 2-3.

Learning goals: technical skills

After successful completion of this lab, you will be able to:
• Explain the fundamental mission of AI for Science, why it is relevant and what is the necessary skill set for
successfully applying it to real-world problems.
• Evaluate the impact of AI for Science on different scientific areas, and explain why it has made an impact on some
areas and not much on others.
• Review different AI for Science technologies, with a focus on their weaknesses/strengths across disciplines, such as
material design and discovery.
• Analyze the potential of AI in real-world experimental settings, such as applying Bayesian optimization for design
of experiments.
• Understand necessary components for building generative and predictive AI tools for different scientific questions.

Learning goals: soft skills

Contents
AI for Science is the term loosely used for applying modern artificial intelligence methods to accelerate discovery and
deepen understanding in the natural and applied sciences. It uses generative models to suggest new ideas, neural
networks to replace slow simulations, and Bayesian optimizers to guide experiments as they progress. Together,
these tools greatly reduce the time to advance science and make discoveries. Initial breakthroughs highlight the
magnitude of this paradigm shift: AlphaFold 3 now predicts full protein–nucleic acid–ligand complexes with
near-atomic accuracy, condensing months of wet-lab crystallography into minutes of GPU time. At the laboratory
bench, agentic platforms are beginning to knit simulations, robotic synthesizers and language-model planning into a
continuous, self-refining loop of experimentation.
In this lab, you will study these methods in depth by implementing some of them and running different analyses on
them. You will evaluate your learnings by presenting the method for the class, giving a short demonstration and
delivering a final report.

Prerequisites
Recommended:
The lab requires knowledge in mathematics (e.g., linear algebra, calculus, optimization) and programming (e.g.
Python, especially PyTorch). Knowledge related to modern AI and deep learning, especially in the context of visual
computing or robotics will help with a smooth experience.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 148

MA-INF 4304 Lab Cognitive Robotics

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Sven Behnke Prof. Dr. Sven Behnke, Dr. Raphael Memmesheimer

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Participants acquire practical experience and in-depth knowledge in the design and implementation of perception
and control algorithms for complex robotic systems. In a small group, they analyze a problem, realize a
state-of-the-art solution, and evaluate its performance.

Learning goals: soft skills

Self-competences (time management, goal-oriented work, ability to analyze problems and to find practical solutions),
communication skills (Work together in small teams, oral and written presentation of solutions, critical examination
of implementations)

Contents
Robot middleware (ROS), simultaneous localization and mapping (SLAM), 3D representations of objects and
environments, object detection and recognition, person detection and tracking, action recognition, action planning
and control, mobile manipulation, human-robot interaction.

Prerequisites
Recommended:
At least 1 of the following:
• MA-INF 4113 – Cognitive Robotics
• MA-INF 4114 – Robot Learning

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
• S. Thrun, W. Burgard and D. Fox: Probabilistic Robotics. MIT Press, 2005.
• B. Siciliano, O. Khatib (Eds.): Springer Handbook of Robotics, 2008.
• Selected research papers.
Master Computer Science — Universität Bonn 149

MA-INF 4306 Lab Development and Application of Data Mining and

Learning Systems

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Stefan Wrobel Prof. Dr. Stefan Wrobel
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Students will acquire in-depth knowledge in the design, implementation, and experimental evaluation of machine
learning and data mining systems. They learn how to work with existing state-of-the-art machine learning and data
mining algorithms and apply them to real-world and synthetic datasets, usually extending them for the requirements
of their particular learning/mining task.

Learning goals: soft skills

Ability to analyze problems theoretically and to find creative and practical solutions; to examine one’s solutions and
results critically; to classify one’s own results into the state-of-the-art of the respective area; to prepare readable
documentation of software and research results; to present, defend and discuss design decisions and results in the
team/group and to other students clearly and in accordance with academic standards; to collaborate constructively
with others in small teams over a longer period of time; to aim at long-range goals under limited resources; to work
under pressure.

Contents
Design, adaptation, implementation, and systematic experimental evaluation of specialised data mining and learning
algorithms, from classical to state-of-the-art, from all areas of machine learning and data mining. Search and
optimization algorithms. Common open source libraries for machine learning and data mining.

Prerequisites
Recommended:
Basic notions and algorithms from machine learning and data mining are required. It is recommended to take at
least one of the following courses first:
• MA-INF 4111 – Principles of Machine Learning
• MA-INF 4112 – Algorithms for Data Science

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Forms of media
Computer Software, Documentation, Research Papers.

Literature
The relevant literature will be announced towards the end of the previous semester.
Master Computer Science — Universität Bonn 150

MA-INF 4308 Lab Vision Systems

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every semester

Module coordinator Lecturer(s)

Prof. Dr. Sven Behnke Prof. Dr. Sven Behnke
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Students will acquire knowledge of the design and implementation of parallel algorithms on GPUs. They will apply
these techniques to accelerate standard machine learning algorithms for data-intensive computer vision tasks.

Learning goals: soft skills

Self-competences (time management, goal-oriented work, ability to analyze problems and to find practical solutions),
communication skills (Work together in small teams, oral and written presentation of solutions, critical examination
of implementations)

Contents
Basic matrix and vector computations with GPUs (CUDA). Classification algorithms, such as multi-layer
perceptrons, support-vector machines, k-nearest neighbors, linear-discriminant analysis. Image preprocessing and
data handling. Quantitative performance evaluation of learning algorithms for segmentation and categorization.

Prerequisites
Recommended:
At least 1 of the following:
MA-INF 2201 - Computer Vision
MA-INF 4111 – Intelligent Learning and Analysis Systems: Machine Learning
MA-INF 4204 – Technical Neural Nets
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
• R. Szeliski: Computer Vision: Algorithms and Applications, Springer 2010.
• C. M. Bishop: Pattern Recognition and Machine Learning, Springer 2006.
• NVidia CUDA Programming Guide, Version 4.0, 2011.
Master Computer Science — Universität Bonn 151

MA-INF 4322 Lab Machine Learning on Encrypted Data

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Dr. Michael Nüsken Dr. Michael Nüsken
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

The students will carry out a practical task (project) in the context of Cryptography, including test and
documentation of the implemented software/system.

Learning goals: soft skills

Ability to properly present and defend design decisions, to prepare readable documentation of software; skills in
constructively collaborating with others in small teams over a longer period of time; ability to classify one’s own
results into the state-of-the-art of the resp. area

Contents
With the rise of more and more mechanisms and installations of data science methodology to automatically analyze
large amounts of possibly privacy infringing data we have to carefully understand how to protect our data. Also
more and more fake data shows up and we have to find ways to distinguish faked from trustable data. At the same
time we want to allow insightful research and life-easing analyzes to be possible. This seeming contradiction has lead
to various efforts for unifying both: protecting data and allowing analyzes, at least to some extent and possibly
under some restrictions.
The target of the lab is to understand how computations on encrypted data may work in one particular application
that we are choosing together. Ideally, we can come up with a novel solution for performing an unconsidered
algorithm. We study the tasks and tools, select algorithms, find a protocol, prototype an implemention, perform a
security analysis, present an evaluation.

Prerequisites
Recommended:
Good knowledge in cryptography is vital, e.g. by one or more modules out of:
• MA-INF 1103 - Cryptography,
• MA-INF 1223 - Privacy Enhancing Technologies, and
• MA-INF 1209 - Seminar Advanced Topics in Cryptography.

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 152

MA-INF 4324 Seminar Advanced Topics in Data Science

Workload Credit points Duration Frequency

120 h 4 CP 1 semester at least every 2 years

Module coordinator Lecturer(s)

Prof. Dr. Elena Demidova Prof. Dr. Elena Demidova
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Enhanced and in-depth knowledge in specialized topics in the data science, including understanding of the data
science process, statistical and machine learning-based data analytics methods, specialized data representation
techniques, evaluation methods, and domain-specific applications.

Learning goals: soft skills

Acquire the competence to independently search for and study state-of-the-art scientific literature in depth, read
critically, identify the most relevant content, and assess research results in the context of the corresponding research
area; to discuss research results with a knowledgeable scientific audience; to present prior work by others in writing
and in presentations with visual media in a way that adheres to academic standards, that is well-structured and
didactically effective, and that motivates the audience to participate; to formulate and accept criticism; to manage
one’s time with relatively open assignments and long-ranging deadlines.

Contents
Statistical and machine learning-based methods of data analytics, including typical steps of the data science process:
data generation, integration, cleaning, exploration, modelling and evaluation. Specialized data representation and
analytics methods for selected data types and applications in specific domains.

Prerequisites
Recommended:
MA-INF 4328 - Spatio-Temporal Data Analytics

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
Relevant literature will be announced at the beginning of the seminar
Master Computer Science — Universität Bonn 153

MA-INF 4325 Lab Data Science in Practice

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Elena Demidova Dr. Rajjat Dadwal

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

This module concentrates on practical experience in data analytics. Participants acquire basic knowledge and
practical experience in the design and implementation of data science workflows for specific data types and
applications.

Learning goals: soft skills

• Communication skills: the ability to work in teams.
• Self-competences: the ability to analyse problems and find practical solutions. Time management, creativity,
presentation of results.

Contents
Practical application of statistical and machine learning-based methods to solve data analytics problems on
real-world datasets and evaluate proposed solutions.

Prerequisites
Recommended:
MA-INF 4328 - Spatio-Temporal Data Analytics

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 154

MA-INF 4326 Explainable AI and Applications

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Dr. Rafet Sifa Prof. Dr. Rafet Sifa, Dr. Lorenz Sparrenberg

Programme Mode Semester

M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

• Know the dual-model functioning of the human mind, and two main AI paradigms
• Develop white-box neural AI systems
• Understand the problems and limitations of Blackbox Deep-Learning systems, and Know the state-of-the-art
Methods for Interpreting Deep-Learning systems (XAI)

Learning goals: soft skills

• Know System 1 and 2 of the mind, prons and cons of symbolic AI and connectionist AI
• Develop neural-geometric systems that have both good features of symbolic AI and connectionist AI
• Know the limitation of famous Deep-Learning systems, such as GPT3, self-driving. Know standard methods to
explore the explainability of Deep-Learning systems

Contents
1. Introduction: fates of large Deep-Learning systems, e.g. Watson, GPT, self-driving cars
2. Dual-system theories (System 1 and 2), nine laws of cognition, criteria of semantic models
3. The target and the state-of-art methods of XAI
4. Neural-symbolic AI
5. Cognitive maps, Collages, Mental Spatial Representation, Events
6. Qualitative Spatial Representation and Reasoning
7. Rotating Sphere Embedding: A New Wheel for Neural-Symbolic Unification
8. Neural Syllogistic Reasoning
9. Recognizing Variable Environments
10. Humor Understanding
11. Rotating Spheres as building-block semantic components for Language, Vision, and Action

Prerequisites
none
Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to four students. A total
of 50% of the points must be achieved.
Master Computer Science — Universität Bonn 155

Literature
• Kahneman, D. (2011). Thinking fast and slow. Farrar, Straus and Giroux.
• Gaedenfors, P. (2017). The Geometry of Meaning. MIT Press.
• Attardo, Hempelmann, Maio (2003). Script Oppositions and Logical Mechanisms: Modeling Incongruities and
their Resolutions, HUMOR 15(1)3–46
• Tversky, B. (2019). Mind in Motion. Basic Books, New York.
• Dong, et al. (2020). Learning Syllogism with Euler Neural-Networks. arXiv:2007.07320
• Dong, T. (2021). A Geometric Approach to the Unification of Symbolic Structure and Neural Networks. Springer.
• Knauff and Spohn (2021). Handbook of Rationality. MIT Press, Cambridge, MA, USA.
• Samek [Link]. (2019), Explainable AI: Interpreting, Explaining and Visualizing Deep Learning. Springer.
• Greg Dean (2019). Step by Step to Stand-Up Comedy (Revised Edition). ISBN: 978-0-9897351-7-9
Master Computer Science — Universität Bonn 156

MA-INF 4327 Lab Biomedical Data Science

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Holger Fröhlich Prof. Dr. Holger Fröhlich

Programme Mode Semester

M. Sc. Computer Science Optional 3.

Learning goals: technical skills

The students will carry out a practical task (project) in the context of biomedical data science, including test and
documentation of the implemented software/system.

Learning goals: soft skills

Ability to properly present and defend design decisions, to prepare readable documentation of software; skills in
constructively collaborating with others in small teams over a longer period of time; ability to classify ones own
results into the state-of-the-art of the resp. area

Contents
Varying selected topics close to current research in the area of biomedical data science.

Prerequisites
none
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 157

MA-INF 4328 Spatio-Temporal Data Analytics

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Elena Demidova Dr. Rajjat Dadwal, Prof. Dr. Elena Demidova

Programme Mode Semester

M. Sc. Computer Science Optional 1. or 2.

Learning goals: technical skills

This module introduces the students to the advanced methods, data structures, and data analytics algorithms for
spatio-temporal data. At the end of the module, the students will be capable of choosing appropriate data
representations, data structures and algorithms for specific applications and correctly applying relevant statistical
and machine learning-based data analytics procedures.

Learning goals: soft skills

Communication skills: oral and written presentation and discussion of solutions. Self-competences: the ability to
analyze and solve problems.

Contents
The module topics include data structures, data representation and analysis methods, and algorithms that enable
analyzing spatio-temporal data and building predictive models effectively and effectively. Furthermore, we will study
the corresponding evaluation techniques and novel applications.

Prerequisites
Recommended:
Basic knowledge of data science and machine learning; programming skills. Recommended reading:
• Sarah Boslaugh. Statistics in a Nutshell. A Desktop Quick Reference, O’Reilly Media, Inc., 2nd Edition, (2012).
• Ethem Alpaydin. Machine Learning. The MIT Press (2021).

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (90 minutes)

Ungraded coursework (required for admission to the exam)

The completion of regularly provided exercise sheets. The work can be done in groups of up to three, four or five
students, depending on the total number of students taking the course. A total of 50% of the points must be
achieved. For 80% of the exercise sheets, 40% of the points must be achieved for each sheet. Each student must
present a solution to an exercise in the exercise sessions once.
Master Computer Science — Universität Bonn 158

MA-INF 4329 Seminar Biological Intelligence

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Dr. Dominik Bach Prof. Dr. Dr. Dominik Bach
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Ability to understand new research results presented in original scientific papers.

Learning goals: soft skills

Communication skills: oral and written presentation of scientific content. Self-competences: the ability to analyze
problems, time management, creativity

Contents
Humans and other animals outperform artificial agents in various tasks and domains. This includes but is not
limited to: learning and planning in unstructured domains; learning from sparse data, observation, and play;
generalisation and transfer; causal reasoning; intuitive physics and psychology; language use; any time planning;
continuous planning; spatial navigation; dynamic memory and active forgetting. This seminar provides background
on some of the underlying biological skills, and computational theories that seek to explain them. We will discuss
implications for designing and/or constraining artificial agents.

Prerequisites
none
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 10 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 159

MA-INF 4330 Lab Explainable AI and Applications

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Rafet Sifa Prof. Dr. Rafet Sifa, Dr. Lorenz Sparrenberg

Programme Mode Semester

M. Sc. Computer Science Optional 2-4.

Learning goals: technical skills

Independent Research: Students will select a research paper focused on representation learning, replicate its findings,
and use techniques from the "Explainable AI and Applications" course to deepen their understanding of the concepts
and potentially enhance the results. This process also teaches students to manage and complete a project within a
defined scope and timeline.
Practical Application: The lab emphasizes the application of theoretical knowledge to real-world problems,
encouraging deeper understanding and innovation. Students will become familiar with external research, apply, and
adapt the relevant research code to their projects.
Communication Skills: Students will develop their ability to present complex ideas clearly and effectively through
structured presentations and written reports. The course also covers scientific writing, literature review, proper
citation, and best practices in academic research.

Learning goals: soft skills

Contents
The lab focuses on enhancing students’ understanding of Explainable AI and its applications through hands-on
exercises and active participation in presentation meetings. Students explore recent research on the topic of latent
representations (e. g. text or image embeddings, sentiment analysis) aiming to reproduce existing research. Then,
they apply techniques learned in the lecture “Explainable AI and Applications” (e. g. neurosymbolic representation
learning) to get a better understanding of these representations. The results will be presented and discussed in a
presentation as well as in a student paper (5-8 pages, given template). There is an opportunity to publish excellent
ideas that go beyond the state of the art and brilliant experimental results.
Schedule:
1. organization (April)
2. presentation of the research idea and its application (1 week later)
3. midterm presentation of results (June)
4. final presentation (September)
5. Student paper (September)

Prerequisites
Recommended:
Basic knowledge of machine learning, and pattern recognition, Python programming

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Intermediate presentation (25%), final presentation (25%), student paper (50%)

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 160

Literature
• Topic dependent to be researched by the student.
• Lecture notes of the instructors (Explainable AI and Applications by Dr. Tiansi Dong, WS23/24)
Master Computer Science — Universität Bonn 161

MA-INF 4331 Lab Perception and Learning for Robotics

Workload Credit points Duration Frequency

270 h 9 CP 1 semester at least every year

Module coordinator Lecturer(s)

JProf. Dr. Hermann Blum JProf. Dr. Hermann Blum
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Participants learn how to practically approach a robot perception problem. They learn how to critically read a
research paper, how to conduct experiments in the context of robot perception, and how to report and present
scientific findings.

Learning goals: soft skills

Ability to analyze problems theoretically and to find creative and practical solutions; to examine one’s solutions and
results critically; to classify one’s own results into the state-of-the-art of the respective area; to prepare readable
documentation of software and research results; to present, defend and discuss design decisions and results in the
team/group and to other students clearly and in accordance with academic standards; to collaborate constructively
with others in small teams over a longer period of time; to aim at long-range goals under limited resources; to work
under pressure.

Contents
In small groups, students apply their knowledge of robot perception, deep learning, and computer vision to a novel
problem. They analyze the problem, read into relevant literature, propose and implement a solution, and empirically
test it. They then refine their approach based on an analysis of the experimental outcomes. The course projects are
related to one of multiple of the following topics: Robot localization, planning, navigation, manipulation; Practical
aspects of Deep Learning; Sensor models, calibration, capture, processing. Software deployment.

Prerequisites
Recommended:
Students are expected to have general programming skills and prior experience with python. Students will need to
operate linux terminal systems such as the university’s GPU cluster.
It is recommended to first take two of the following modules:
• MA-INF 2201 Computer Vision
• MA-INF 2213 Advanced Computer Vision
• MA-INF 4113 Cognitive Robotics

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
• S. Thrun, W. Burgard and D. Fox: Probabilistic Robotics. MIT Press, 2005
• I. Goodfellow, Y. Bengio and A. Courville: Deep Learning. MIT Press, 2016
• Per-project assigned literature
Master Computer Science — Universität Bonn 162

MA-INF 4332 Seminar Large Language Models

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Lucie Flek Prof. Dr. Lucie Flek
Programme Mode Semester
M. Sc. Computer Science Optional 2-4.

Learning goals: technical skills

Learning goals: soft skills

Contents
Large Language Models (LLMs), such as GPT-4, Gemini, and their successors, have had an enormous impact on
various domains, including natural language processing, machine learning, and artificial intelligence. These models
have redefined what’s possible in applications such as text generation, translation, summarization, sentiment
analysis, and more. The aim of this seminar is to explore cutting-edge research, insights, and trends in the field of
LLMs, such as:
• hallucination reduction and factual grounding
• explainability, reasoning, faithfulness
• safety, toxicity, fairness and bias
• social and moral alignment of LLMs
• style control and personalization
• sustainability, compression, model size reduction, knowledge distillation
• multilinguality and multimodality
• LLMs as planning agents
• and more
Prerequisites
none
Remarks
At most 35 participants

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 35 2 30 T / 90 S 4

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
• Bommasani, Rishi: On the opportunities and risks of foundation models
• Devlin, Jacob, et al.: Bert: Pre-training of deep bidirectional transformers for language understanding
• Brown, Tom, et al.: Language models are few-shot learners
• WX Zhao, et al.: A survey of large language models
• Yang, Jingfeng, et al.: Harnessing the power of LLMs in practice: A survey on ChatGPT and beyond
Master Computer Science — Universität Bonn 163

MA-INF 4333 Geometric Deep Learning

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Jun. Prof. Dr. Zorah Lähner Prof. Dr. Zorah Lähner
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

• Understanding advanced topics in the design of neural networks using geometric data
• Mathematical modelling of invariances and non-Euclidean domains in deep learning and guarantees that can be
derived from these
• Gain an overview of practical applications in which this theory can be applied

Learning goals: soft skills

• Problem solving skills: ability to identify and utilize analogies between new problems and previously seen ones
• Analytical and abstract thinking: develop a general intuition of computational problems, being able to adopt
different perspectives of particular concepts

Contents
This lecture will cover advanced topics in deep learning focusing on theory related to geometric data and the
incorporation of invariances in network architectures. Topics include, among others, permutation invariance,
differential geometry, the curse of dimensionality, neural fields and physics-informed neural networks. Students will
learn how to process a variety of geometric data structures and implement deep learning algorithms on these related
to applications in visual computing, physics and graph processing.

Prerequisites
Recommended:
Students are recommended to have basic knowledge about deep learning and computer vision, for example gained in:
• MA-INF 4111 Principles of Machine Learning,
• MA-INF 2201 Computer Vision, or
• MA-INF 2222 Visual Data Analysis,
and proficiency in python.

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam (120 minutes)

Ungraded coursework (required for admission to the exam)

none
Master Computer Science — Universität Bonn 164

MA-INF 4334 Computational neuroscience: cognition and behaviour

Workload Credit points Duration Frequency

180 h 6 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Dr. Dominik Bach Prof. Dr. Dr. Dominik Bach
Programme Mode Semester
M. Sc. Computer Science Optional 2-3.

Learning goals: technical skills

• Conceptual knowledge and mathematical understanding of common behavioural and cognitive models from
computational neuroscience
• Knowledge of common experimental methods used to develop and disambiguate such models
• Basic knowledge of fundamentals in neuroscience, cognitive/perceptual psychology and microeconomics
• In the exercises, students will learn to implement models and how to use them as benchmarks for bottom-up
computational neuroscience models, and for automatic signature-testing of AI algorithms

Learning goals: soft skills

• Teamwork (exercises)
• Oral presentation in front of audience (exercises)

Contents
The two dominant paradigms in computational neuroscience are bottom-up (starting from the spontaneous
behaviour of constituent elements of the nervous system) and top-down (starting from known functions of biological
agents). This lecture introduces important topdown models of behaviour and cognition from three perspectives:
computational (problem definition and optimal solutions), algorithmic (rational/engineering/descriptive solutions)
and implementation (neural hardware). The lecture covers the following domains:
• decision-making with noisy information (value-based, time-integrated, multi-channel, sequential)
• information representation under resource constraints
• memory formation and storage in biological neural networks
• movement planning
• spatial navigation

Prerequisites
Recommended:
Recommended one out of:
• MA-INF 4113 Cognitive Robotics
• MA-INF 4114 Robot Learning
• MA-INF 4215 Humanoid Robotics
• MA-INF 4235 Reinforcement Learning

Course meetings
Teaching format Group size h/week Workload[h] CP T = face-to-face teaching
Lecture 2 30 T / 45 S 2.5 S = independent study
Exercises 2 30 T / 75 S 3.5

Graded exams
Written exam
Ungraded coursework (required for admission to the exam)
The completion of regularly provided exercise sheets. The work can be done in groups of up to four students. A total
of 50% of the points must be achieved. Each group must present a solution to an exercise in the exercise sessions
once.
Master Computer Science — Universität Bonn 165

MA-INF 4335 Lab AI Alignment

Workload Credit points Duration Frequency

270 h 9 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Lucie Flek Prof. Dr. Lucie Flek
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

Through tutorials and a final project, you will gain hands-on experience in AI alignment techniques and have this
chance to apply this knowledge to various interesting projects. Students will collaborate in small teams (a group of
3-4 students) and implement small research projects over the course of the term, advised by a researcher from the
CAISA lab. Students will learn to reproduce important results from the field, study the scientific literature, generate
and implement their own research ideas, and present their results as a presentation and as a paper.

Learning goals: soft skills

Contents
As AI systems such as Large Language Models become increasingly capable and start to be used in high-stakes
scenarios, ensuring that they act safely is gaining importance. The research field of AI alignment studies methods to
align the behavior and values of AI systems with the user and broader society in a robust, scalable, and interpretable
way. The aim of this course is to explore cutting-edge research, insights, and trends in the field of AI alignment.
Schedule:
• Week 0: Organization meeting
• Week 1-5: Lectures and programming exercises
• Week 6: Presentation of project ideas
• Week 12: midterm presentation of results
• Final presentation
• Student paper
Concrete research topics are 1) value alignment, 2) emergent misalignment, 3) scalable oversight, and 4) mechanistic
interpretability, among others.

Prerequisites
Recommended:
One of the following courses is recommended:
• MA-INF 4115 - Introduction to Natural Language Processing,
• MA-INF 4235 - Reinforcement Learning,
• MA-INF 4204 - Technical Neural Nets.
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Lab 8 4 60 T / 210 S 9

Graded exams
Oral presentation, written report

Ungraded coursework (required for admission to the exam)

Project work; attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).
Master Computer Science — Universität Bonn 166

Literature
• Hendrycks, Dan: Introduction to AI Safety, Ethics, and Society.
• Ouyang, Long et al.: Training language models to follow instructions with human feedback.
• Bowman, Sam et al.: Measuring Progress on Scalable Oversight for Large Language Models.
• Li, Nathaniel et al.: Measuring and Reducing Malicious Use With Unlearning.
• Bricken, Trenton et al.: Towards monosemanticity: Decomposing language models with dictionary learning.
Master Computer Science — Universität Bonn 167

MA-INF 4336 Seminar Selected Topics in Natural Language Processing

Workload Credit points Duration Frequency

120 h 4 CP 1 semester every year

Module coordinator Lecturer(s)

Prof. Dr. Lucie Flek Prof. Dr. Lucie Flek
Programme Mode Semester
M. Sc. Computer Science Optional 2. or 3.

Learning goals: technical skills

In this course, students will learn to engage with technical research papers and other material that study
the state-of-the-art approaches to Natural Language Processing. They will learn to synthesize, critically
assess, and build upon the material by conducting required readings, weekly discussions. In particular:
• to read and examine the scientific literature critically,
• to identify and present examples of relevant real-world application fields of the scientific publication,
• to give a good scientific presentation to peers that is precise, informative, and engaging,
• to summarize and write a concise technical report including a critical discussion of the research paper (with
additional literature research).

Learning goals: soft skills

Contents
In this weekly seminar, students will learn about selected research topics in natural language processing and its
applications. In addition, it is an introduction to reading scientific papers, learning how to give a good presentation,
and how to write a technical report.
To better explore how state-of-the-art research presentations can be constructed, as a part of the seminar students
will also have a chance to attend and discuss presentations of researchers on more senior level presenting their papers
(e.g. guest speakers).

Prerequisites
Recommended:
One of the following courses is recommended (but not obligatory):
• MA-INF 4115 - Introduction to Natural Language Processing,
• MA-INF 4235 - Reinforcement Learning,
• MA-INF 4204 - Technical Neural Nets.
Remarks
At most 20 participants

Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 20 2 30 T / 90 S 4

Graded exams
Oral presentation, written report.

Ungraded coursework (required for admission to the exam)

Attendance in course sessions in accordance with the exam regulations of 2023, § 12(6).

Literature
• Jurafsky & Martin, Speech and Language Processing: [Link] .
Master Computer Science — Universität Bonn 168

5 Master Thesis
MA-INF 0401 30 CP Master Thesis . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 169
MA-INF 0402 2 CP Master Seminar . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
Master Computer Science — Universität Bonn 169

MA-INF 0401 Master Thesis

Workload Credit points Duration Frequency

900 h 30 CP 1 semester every semester

Module coordinator Lecturer(s)

The Examination Board All lecturers of computer science

Programme Mode Semester

M. Sc. Computer Science Compulsory 4.

Learning goals: technical skills

Ability to solve a well-defined, significant research problem under supervision, but in principle independently

Learning goals: soft skills

Ability to write a scientific documentation of considerable length according to established scientific principles of form
and style, in particular reflecting solid knowledge about the state-of-the-art in the field

Contents
Topics of the thesis may be chosen from any of the areas of computer science represented in the curriculum

Prerequisites
Required:
By the examination regulations of 2023, the Master’s thesis project can only commence after 60 credits in other
modules of the programme have been obtained. Before you start on the project, you must obtain the approval of the
exam committee and register the starting date of the project. Please check the website of the examination office for
forms and procedures.

Course meetings
Teaching format Group size h/week Workload[h] CP
Independent 0 900 S 30 T = face-to-face teaching
preparation of a S = independent study
scientific thesis with
individual coaching

Graded exams
Master Thesis
Ungraded coursework (required for admission to the exam)
None
Literature
Individual bibliographic research required for identifying relevant literature (depending on the topic of the thesis)
Master Computer Science — Universität Bonn 170

MA-INF 0402 Master Seminar

Workload Credit points Duration Frequency

60 h 2 CP 1 semester every semester

Module coordinator Lecturer(s)

The Examination Board All lecturers of computer science

Programme Mode Semester

M. Sc. Computer Science Compulsory 4.

Learning goals: technical skills

Knowledge of the state-of-the-art in research in the respective area and how the thesis results relate to that.

Learning goals: soft skills

Ability to identify the most relevant content for a knowledgeable scientific audience; ability to present and defend
one’s work in a presentation with visual media in a way that adheres to academic standards; ability to anticipate,
accept and answer critical questions.

Contents
Topic, scientific context, and results of the master thesis

Prerequisites
Required:
The Master Seminar accompanies the Master Thesis project, see MA-INF 0401 for prequisites.
Recommended:
None
Course meetings
T = face-to-face teaching
Teaching format Group size h/week Workload[h] CP
S = independent study
Seminar 2 30 T / 30 S 2

Graded exams
Oral presentation of final results

Ungraded coursework (required for admission to the exam)

None
Literature
Individual bibliographic research required for identifying relevant literature (depending on the topic of the thesis)