26/01/26, 7:18 PM

Department of Artificial Intelligence and Data Science --

KLVZA
Course Handout
K L Deemed to be University
2025-2026, Even Sem

Course Title : EXPLAINABLE AI AND COMMUNICATIONS

Course Code : 23ACI3202R
L-T-P-S Structure : 3-0-0-0
Pre-requisite :
Credits :3
Course Coordinator : Joe Nithin Veguri
Team of Instructors :
Teaching Associates :
Foundations of XAI, ethical & regulatory considerations, transparency,
trust, limitations of black-box models Conceptual XAI techniques:
model-agnostic vs model-specific approaches, feature importance,
local/global interpretability, post-hoc explanations Applications of XAI
Syllabus :
in healthcare, finance, autonomous systems, NLP, computer vision,
cross-domain impacts on trust and ethics Advanced topics: human-
centered XAI, regulatory compliance, causal models, counterfactual
explanations, interactive explanation systems, emerging research trends
1. Christoph Molnar, Interpretable Machine Learning, 2nd Edition,
2022, [Link] 2. Been Kim
Text Books :
et al., Explainable AI: Interpreting, Explaining and Visualizing Deep
Learning, Springer, 2021
1. Finale Doshi-Velez and Been Kim, Towards a Rigorous Science of
Interpretable Machine Learning, 2017, [Link]
Reference Books :
2. Adadi, A., & Berrada, M., Peeking Inside the Black-Box: A Survey
on Explainable AI (XAI), IEEE Access, 2018
• [Link] •
[Link] •
[Link] •
Web Links : [Link]
and-applications-4518d1022381 •
[Link]
explainable-ai-xai/
• Coursera: Explainable AI: Interpreting Machine Learning Models
([Link] • edX: AI for Everyone
– foundational coverage with ethical context
([Link] • Udemy:
MOOCS :
Interpretable Machine Learning in Python
([Link] •
FutureLearn: Ethics and Law in AI
([Link]
The Advanced Mode Explainable AI and Communications course is
designed to cultivate a deep, conceptual understanding of AI

[Link] Page 1 of 29
 26/01/26, 7:18 PM

interpretability, transparency, ethical principles, and effective

communication of AI insights. It equips students to: 1. Understand
Foundations: Grasp the philosophical, ethical, and regulatory
motivations behind XAI and its communication. 2. Analyze
Techniques: Conceptually explore model-agnostic and model-specific
interpretability methods, feature importance, and post-hoc explanations.
3. Evaluate Applications: Assess the impact of XAI in domains like
Course Rationale : healthcare, finance, autonomous systems, and NLP, focusing on trust,
fairness, societal implications, and clear communication of results. 4.
Explore Advanced Frontiers: Discuss human-centered design, causal
reasoning, counterfactual explanations, and emerging research trends to
foster critical thinking, effective communication, and lifelong learning.
By emphasizing theoretical mastery, reflective thinking, and
communication of insights, the course aligns with your intent to deliver
knowledge that is deep, structured, and enduring, preparing students to
understand, evaluate, and articulate AI systems responsibly and
ethically.
To develop a comprehensive conceptual understanding of Explainable
AI (XAI) and its communication aspects, enabling students to critically
Course Objectives : analyze, evaluate, and convey interpretability methods, ethical
considerations, and domain-specific applications, fostering responsible,
transparent, and human-centric AI reasoning.

COURSE OUTCOMES (COs):

Blooms
CO Taxonomy
Course Outcome (CO) PO/PSO
NO Level
(BTL)
Understand the fundamental principles of explainable AI and
CO1 PO1,PO5,PSO1 2
its importance in ethical AI development.
Apply model-agnostic and interpretable machine learning
CO2 PO1,PO2,PO6,PSO1 3
techniques to real-world datasets.
Critically evaluate explainable AI systems for reliability,
CO3 PO2,PO3,PO5,PSO1 5
trustworthiness, and explanation quality.
Communicate AI decisions effectively using concepts of
CO4 PO3,PO4,PO5,PSO1 5
explainability and ethical AI.

COURSE OUTCOME INDICATORS (COIs)::

Outcome Highest
COI-2 COI-3 COI-4 COI-5
No. BTL
Btl-2
Describe
CO1 2
transparency and
trust requirements
Btl-3 Btl-3
[Link] Page 2 of 29
 26/01/26, 7:18 PM

Demonstrate local Use model-agnostic

CO2 3
and global and model-specific
interpretability XAI tools
Btl-4
Btl-2 Btl-5
Btl-3 Analyze the
Describe the need Critically evaluate
Apply the traits of reliability of
for reliability, explainable AI
reliability, systems on
CO3 5 trustworthiness, and systems for
trustworthiness and application of the
explanation quality reliability,
explanation quality traits of reliability,
in explainable AI trustworthiness, and
in AI systems. trustworthiness, and
systems. explanation quality.
explanation quality.
Btl-5
Btl-2
Btl-3 Evaluate
Explain the need
Apply the concepts Btl-4 communicate AI
for effective
of explainability for Analyze the decisions
CO4 5 communication of
effective effectiveness in AI effectively using
AI decision making
communication in systems. concepts of
in ethical
AI systems. explainability and
conscience.
ethical AI.

PROGRAM OUTCOMES & PROGRAM SPECIFIC OUTCOMES (POs/PSOs)

Po
Program Outcome
No.
Engineering Knowledge:Apply the knowledge of mathematics, science, engineering
PO1
fundamentals, and an engineering specialization to the solution of complex engineering problems.
Problem Analysis: Identify, formulate, review research literature, and analyse complex
PO2 engineering problems reaching substantiated conclusions using first principles of mathematics,
natural sciences and engineering sciences.
Design/Development of Solutions: Design solutions for complex engineering problems and
design system components or processes that meet the specified needs with appropriate
PO3
consideration for the public health and safety, and the cultural, societal, and environmental
considerations
Conduct Investigations of Complex Problems:Use research-based knowledge and research
methods including design of experiments, analysis and interpretation of data, and synthesis of the
PO4 information to provide valid conclusions for complex problems that cannot be solved by
straightforward application of knowledge, theories and techniques applicable to the engineering
discipline.
Modern Tool Usage:Create, select, and apply appropriate techniques, resources, and modern
PO5 engineering and IT tools including prediction and modelling to complex engineering activities
with an understanding of the limitations.
The Engineer and Society:Apply reasoning informed by the contextual knowledge to assess
PO6 societal, health, safety, legal and cultural issues and the consequent responsibilities relevant to the
professional engineering practice.
An ability to design and develop Artificial Intelligence technology into innovative products for
PSO1
solving real world problems

[Link] Page 3 of 29
 26/01/26, 7:18 PM

Lecture Course DELIVERY Plan:

Book
No[CH
Teaching-
[Link]. CO COI Topic No] EvaluationComponents
Learning Methods
[Page
No]

COI- Introduction to AI & End Semester

1 CO1 TB Chalk,LTC,PPT,Talk
2 Subfields Exam,SEM-EXAM1

COI- AI Overview: ML, DL, End Semester

2 CO1 TB Chalk,LTC,PPT,Talk
2 NLP, CV Exam,SEM-EXAM1

COI- Scope and End Semester

3 CO1 TB Chalk,LTC,PPT,Talk
2 Applications of AI Exam,SEM-EXAM1

Importance of
COI- ALM,End Semester
4 CO1 Explainability in AI TB Chalk,LTC,PPT,Talk
2 Exam,SEM-EXAM1
Systems

Ethical Considerations:
COI- End Semester
5 CO1 Bias, Fairness, TB Chalk,LTC,PPT,Talk
2 Exam,SEM-EXAM1
Transparency

COI- Ethical Considerations: End Semester

6 CO1 TB Chalk,LTC,PPT,Talk
2 Privacy, Accountability Exam,SEM-EXAM1

Regulatory
COI- End Semester
7 CO1 Touchpoints: GDPR, TB Chalk,LTC,PPT,Talk
2 Exam,SEM-EXAM1
Auditability

COI- Case Studies: End Semester

8 CO1 TB Chalk,LTC,PPT,Talk
2 Healthcare AI Exam,SEM-EXAM1

COI- Case Studies: Finance ALM,End Semester

9 CO1 TB Chalk,LTC,PPT,Talk
2 AI Exam,SEM-EXAM1

End Semester
COI- Case Studies: Policing Exam,Home
10 CO1 TB Chalk,LTC,PPT,Talk
2 AI Assignment,SEM-
EXAM1

TB1,
COI- Interpretable Machine Ch.5, End Semester
11 CO2 LTC,PPT,Talk
2 Learning Techniques pp.175- Exam,SEM-EXAM1
190

[Link] Page 4 of 29
 26/01/26, 7:18 PM

TB1,
COI- Linear Models and Ch.5, End Semester
12 CO2 LTC,PPT,Talk
3 Logistic Regression pp.190- Exam,SEM-EXAM1
205

COI- Decision Trees, Rule- End Semester

13 CO2 TB Chalk,LTC,PPT,Talk
3 Based Systems Exam,SEM-EXAM1

TB1,
COI- Feature Engineering Ch.5, ALM,End Semester
14 CO2 LTC,PPT,Talk
3 for Interpretability pp.175- Exam,SEM-EXAM1
190

TB1,
Model-Agnostic
COI- Ch.6, End Semester
15 CO2 Explanation Methods: LTC,PPT,Talk
3 pp.250- Exam,SEM-EXAM1
LIME, Anchors
265

Model-Agnostic
COI- End Semester
16 CO2 Explanation Methods: TB Chalk,LTC,PPT,Talk
3 Exam,SEM-EXAM1
SHAP, PDP

TB1,
Advantages,
COI- Ch.5, End Semester
17 CO2 Limitations, Use-Case LTC,PPT,Talk
3 pp.190- Exam,SEM-EXAM1
Selection
205

TB1,
Local & Global
COI- Ch.4, End Semester
18 CO2 Methods: Perturbation- LTC,PPT,Talk
2 pp.150- Exam,SEM-EXAM1
Based Techniques
170

TB1,
Local & Global
COI- Ch.5, ALM,End Semester
19 CO2 Methods: Case LTC,PPT,Talk
3 pp.190- Exam,SEM-EXAM1
Examples
205

TB1, End Semester

COI- Hands-on Mini- Ch.5, Exam,Home
20 CO2 LTC,PPT,Talk
3 Projects (conceptual) pp.175- Assignment,SEM-
190 EXAM1

Incorporating
COI- End Semester
21 CO3 Explainability: Design TB Chalk,LTC,PPT,Talk
4 Exam,SEM-EXAM2
Phase

Incorporating
22 CO3 COI- Explainability: TB Chalk,LTC,PPT,Talk End Semester

[Link] Page 5 of 29
 26/01/26, 7:18 PM

4 Training Phase Exam,SEM-EXAM2

Incorporating
COI- End Semester
23 CO3 Explainability: Post- TB Chalk,LTC,PPT,Talk
5 Exam,SEM-EXAM2
Deployment

TB2,
Monitoring and
COI- Ch.7, ALM,End Semester
24 CO3 Maintaining LTC,PPT,Talk
5 pp.450- Exam,SEM-EXAM2
Explanations
470

TB3,
Project-Based
COI- Ch.4, End Semester
25 CO3 Exercises: Lifecycle LTC,PPT,Talk
5 pp.475- Exam,SEM-EXAM2
Integration
495

Perturbation-Based
COI- Attribution: End Semester
26 CO3 TB Chalk,LTC,PPT,Talk
3 LIME/Anchors/SHAP Exam,SEM-EXAM1
(deeper)

Propagation-Based
COI- End Semester
27 CO3 Attribution: Relevance TB Chalk,LTC,PPT,Talk
3 Exam,SEM-EXAM1
Propagation

TB2,
Propagation-Based
COI- Ch.4, End Semester
28 CO3 Attribution: Saliency LTC,PPT,Talk
3 pp.210- Exam,SEM-EXAM2
Methods
225

TB1,
Concept-Based
COI- Ch.8, ALM,End Semester
29 CO3 Explainability: LTC,PPT,Talk
4 pp.520- Exam,SEM-EXAM2
Net2Vec, T-CAV, ACE
540

TB1, End Semester

Interpretable
COI- Ch.9, Exam,Home
30 CO3 Architectures: CBMs LTC,PPT,Talk
4 pp.545- Assignment,SEM-
and Variants
565 EXAM2

Interpretable
COI- End Semester
31 CO4 Architectures: Concept TB LTC,PPT,Talk
4 Exam,SEM-EXAM2
Whitening, SENNs

Neurosymbolic TB1,
32 CO4 COI- Methods: Ch.9, LTC,PPT,Talk End Semester
4 DeepProbLog, Neural pp.545- Exam,SEM-EXAM2

[Link] Page 6 of 29
 26/01/26, 7:18 PM

Reasoners 565

Sample-Based
COI- End Semester
33 CO4 Explanations: Influence TB Chalk,LTC,PPT,Talk
3 Exam,SEM-EXAM1
Functions

Sample-Based
COI- ALM,End Semester
34 CO4 Explanations: TB Chalk,LTC,PPT,Talk
3 Exam,SEM-EXAM2
ProtoPNets

Counterfactual
COI- Explanations End Semester
35 CO4 TB Chalk,LTC,PPT,Talk
2 (conceptual & causal Exam,SEM-EXAM1
link)

Case Studies and TB3,

COI- Projects on Building Ch.4, End Semester
36 CO4 LTC,PPT,Talk
4 Explainable AI pp.495- Exam,SEM-EXAM2
Systems 515

COI- Metrics for Evaluating ALM,End Semester

37 CO4 TB Chalk,LTC,PPT,Talk
5 Explanation Quality Exam,SEM-EXAM2

End Semester
COI- Stability, Robustness, Exam,Home
38 CO4 TB Chalk,LTC,PPT,Talk
5 Usability Metrics Assignment,SEM-
EXAM2

COI- Challenges: Accuracy End Semester

39 CO4 TB LTC,PPT,Talk
4 vs Interpretability Exam,SEM-EXAM2

Lecture Session wise Teaching – Learning Plan

SESSION NUMBER : 1

Session Outcome: 1 STUDENTS SHELL BE ABLE TO UNDERSTAND Introduction to AI & Subfields

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 ATTENDANCE 1 Talk APPLICABLE
---
--- NOT
20 Introduction to AI & Subfields 2 PPT APPLICABLE
---

[Link] Page 7 of 29
 26/01/26, 7:18 PM

--- NOT
20 Introduction to AI & Subfields 2 PPT
APPLICABLE
---
--- NOT
5 SUMMERY 2 Talk APPLICABLE
---

SESSION NUMBER : 2

Session Outcome: 1 STUDENT ABLE TO UNDERSTAND AI CONCEPTS: ML, DL, NLP, CV

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 ATTENDANCE 1 Chalk APPLICABLE
---
--- NOT
20 AI Overview: ML, DL, NLP, CV 2 PPT APPLICABLE
---
--- NOT
20 AI Overview: ML, DL, NLP, CV 2 PPT APPLICABLE
---
--- NOT
5 SUMMERY 2 Talk APPLICABLE
---

SESSION NUMBER : 3

Session Outcome: 1 STUDENT ABLE TO UNDERSTAND Scope and Applications of AI

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 ATTENDANCE 1 Talk APPLICABLE
---
--- NOT
20 Scope and Applications of AI 2 PPT APPLICABLE
---
--- NOT
20 Scope and Applications of AI 2 PPT APPLICABLE
---
--- NOT
5 SUMMERY 2 Talk APPLICABLE
---

[Link] Page 8 of 29
 26/01/26, 7:18 PM

SESSION NUMBER : 4

Session Outcome: 1 Students shall understand the Importance of Explainability in AI Systems

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
20 Importance of Explainability in AI Systems 2 PPT APPLICABLE
---
One minute
20 Importance of Explainability in AI Systems 2 PPT
paper
--- NOT
5 Summary 2 Talk APPLICABLE
---

SESSION NUMBER : 5

Session Outcome: 1 STUDENT ABLE TO UNDERSTAND Ethical Considerations: Bias, Fairness,

Transparency

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 ATTENDANCE 1 Talk APPLICABLE
---
--- NOT
20 Ethical Considerations: Bias, Fairness, Transparency 2 PPT APPLICABLE
---
--- NOT
20 Ethical Considerations: Bias, Fairness, Transparency 2 PPT APPLICABLE
---
--- NOT
5 SUMMERY 2 Talk APPLICABLE
---

SESSION NUMBER : 6

Session Outcome: 1 STUDENT TO UNDERSTAND Ethical Considerations: Privacy, Accountability

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods

[Link] Page 9 of 29
 26/01/26, 7:18 PM

--- NOT
5 ATTENDANCE 1 Talk APPLICABLE
---
--- NOT
20 Ethical Considerations: Privacy, Accountability 2 PPT APPLICABLE
---
--- NOT
20 Ethical Considerations: Privacy, Accountability 2 PPT APPLICABLE
---
--- NOT
20 SUMMERY 2 Talk APPLICABLE
---

SESSION NUMBER : 7

Session Outcome: 1 STUDENT TO UNDERSTAND Regulatory Touchpoints: GDPR, Auditability

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 ATTENDANCE 1 Talk APPLICABLE
---
--- NOT
20 Regulatory Touchpoints: GDPR, Auditability 2 PPT APPLICABLE
---
--- NOT
20 Regulatory Touchpoints: GDPR, Auditability 2 PPT APPLICABLE
---
--- NOT
5 SUMMERY 2 Talk APPLICABLE
---

SESSION NUMBER : 8

Session Outcome: 1 STUDENT TO UNDERSTAND Case Studies: Healthcare AI

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 ATTENDANCE 1 Talk APPLICABLE
---
--- NOT
20 Case Studies: Healthcare AI 2 PPT APPLICABLE
---
--- NOT

[Link] Page 10 of 29
 26/01/26, 7:18 PM

20 Case Studies: Healthcare AI 2 PPT APPLICABLE

---
--- NOT
5 SUMMERY 2 Talk APPLICABLE
---

SESSION NUMBER : 9

Session Outcome: 1 Students shall remember the Case Studies: Finance AI

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
20 Case Studies: Finance AI 1 PPT APPLICABLE
---
Think / Pair /
20 Case Studies: Finance AI 1 PPT
Share
--- NOT
5 Summary 2 Talk APPLICABLE
---

SESSION NUMBER : 10

Session Outcome: 1 Students shall remember Case Studies: Policing AI

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
20 Case Studies: Policing AI 1 PPT APPLICABLE
---
--- NOT
20 Case Studies: Policing AI 1 PPT APPLICABLE
---
--- NOT
5 Summary 2 Talk APPLICABLE
---

SESSION NUMBER : 11

[Link] Page 11 of 29
 26/01/26, 7:18 PM

Session Outcome: 1 Student shall understand techniques that make machine learning models inherently
interpretable and easier to explain.

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
40 Interpretable Machine Learning Techniques 2 PPT APPLICABLE
---
--- NOT
5 Summary 1 Talk APPLICABLE
---

SESSION NUMBER : 12

Session Outcome: 1 Student shall understand the interpretability of linear models and logistic regression.

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
40 Linear Models and Logistic Regression 3 PPT APPLICABLE
---
--- NOT
5 Summary 1 Talk APPLICABLE
---

SESSION NUMBER : 13

Session Outcome: 1 STUDENT TO UNDERSTAND Decision Trees, Rule-Based Systems

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 ATTENDANCE 1 Talk APPLICABLE
---
--- NOT
40 Decision Trees, Rule-Based Systems 3 PPT APPLICABLE
---
--- NOT

[Link] Page 12 of 29
 26/01/26, 7:18 PM

5 SUMMARY 2 Talk APPLICABLE

---

SESSION NUMBER : 14

Session Outcome: 1 Students shall enhance model transparency by designing and selecting features that
improve interpretability and explanatory clarity

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
25 Feature Engineering for Interpretability 3 PPT APPLICABLE
---
Inquiry Based
20 ALM on Feature Engineering for Interpretability 3 Talk
Learning

SESSION NUMBER : 15

Session Outcome: 1 Student shall understand how LIME and Anchors provide simple, model-agnostic
local explanations.

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
20 LIME 3 PPT APPLICABLE
---
--- NOT
20 Anchors 3 PPT APPLICABLE
---
--- NOT
5 Summary 1 Talk APPLICABLE
---

SESSION NUMBER : 16

Session Outcome: 1 STUDENT TO UNDERSTAND Model-Agnostic Explanation Methods: SHAP, PDP

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods

[Link] Page 13 of 29
 26/01/26, 7:18 PM

--- NOT
5 ATTENDANCE 1 Talk APPLICABLE
---
--- NOT
20 Model-Agnostic Explanation Methods: SHAP, PDP 3 PPT APPLICABLE
---
--- NOT
20 Model-Agnostic Explanation Methods: SHAP, PDP 3 PPT APPLICABLE
---
--- NOT
5 SUMMARY 2 Talk APPLICABLE
---

SESSION NUMBER : 17

Session Outcome: 1 Students shall evaluate the advantages and limitations of XAI methods to guide
effective use-case selection

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
20 Advantages, Limitations 3 PPT APPLICABLE
---
--- NOT
20 Use-Case Selection 3 PPT APPLICABLE
---
--- NOT
5 Summary 1 Talk APPLICABLE
---

SESSION NUMBER : 18

Session Outcome: 1 Students shall explain model behavior by applying perturbation-based local and
global techniques to assess feature influence on predictions

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
Local & Global Methods: Perturbation-Based
40 2 PPT APPLICABLE
Techniques
[Link] Page 14 of 29
 26/01/26, 7:18 PM

---
--- NOT
5 Summary 1 Talk APPLICABLE
---

SESSION NUMBER : 19

Session Outcome: 1 Students shall illustrate how local and global explanation methods clarify model
behavior through practical, real-world examples

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
25 Local & Global Methods: Case Examples 3 PPT APPLICABLE
---
Reflective
20 ALM on Local & Global Methods: Case Examples 1 Talk
Learning

SESSION NUMBER : 20

Session Outcome: 1 Students shall apply explainable AI concepts through practical mini-projects that
reinforce understanding via implementation

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
40 Hands-on Mini-Projects (conceptual) 3 Talk APPLICABLE
---
--- NOT
5 Summary 1 Talk APPLICABLE
---

SESSION NUMBER : 21

Session Outcome: 1 Students shall analyze Incorporating Explainability: Design Phase

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods

[Link] Page 15 of 29
 26/01/26, 7:18 PM

--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
20 Incorporating Explainability: Design Phase 4 PPT APPLICABLE
---
--- NOT
20 Incorporating Explainability: Design Phase 4 PPT APPLICABLE
---
--- NOT
5 Summary 2 Talk APPLICABLE
---

SESSION NUMBER : 22

Session Outcome: 1 Student shall Analyze the Incorporating Explainability: Training Phase

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
20 Incorporating Explainability: Training Phase 4 PPT APPLICABLE
---
--- NOT
20 Incorporating Explainability: Training Phase 4 PPT APPLICABLE
---
--- NOT
5 Summary 2 Talk APPLICABLE
---

SESSION NUMBER : 23

Session Outcome: 1 Student shall evaluate the Incorporating Explainability: Post-Deployment

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
20 Incorporating Explainability: Post-Deployment 4 PPT APPLICABLE
---
--- NOT

[Link] Page 16 of 29
 26/01/26, 7:18 PM

20 Incorporating Explainability: Post-Deployment 4 PPT APPLICABLE

---
--- NOT
5 Summary 2 Talk APPLICABLE
---

SESSION NUMBER : 24

Session Outcome: 1 Student shall learn how to monitor and maintain explanations to ensure consistent,
reliable, and up-to-date model interpretability over time.

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
20 Monitoring and Maintaining Explanations 5 PPT APPLICABLE
---
Focused
25 ALM on Monitoring and Maintaining Explanations 5 Talk
listing

SESSION NUMBER : 25

Session Outcome: 1 Student shall understand how to integrate explainability throughout the AI project
lifecycle using practical, project-based exercises.

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
40 Project-Based Exercises: Lifecycle Integration 5 PPT APPLICABLE
---
--- NOT
5 Summary 1 Talk APPLICABLE
---

SESSION NUMBER : 26

Session Outcome: 1 STUDENT ABLE TO UNDERSTAND Perturbation-Based Attribution:

LIME/Anchors/SHAP (deeper)

Teaching- Active
Time(min) Topic BTL Learning Learning

[Link] Page 17 of 29
 26/01/26, 7:18 PM

Methods Methods
--- NOT
5 ATTENDANCE 1 Talk APPLICABLE
---
--- NOT
Perturbation-Based Attribution:
20 3 PPT APPLICABLE
LIME/Anchors/SHAP (deeper)
---
--- NOT
Perturbation-Based Attribution:
20 3 PPT APPLICABLE
LIME/Anchors/SHAP (deeper)
---
--- NOT
5 SUMMARY 2 Talk APPLICABLE
---

SESSION NUMBER : 27

Session Outcome: 1 STUDENT ABLE TO UNDERSTAND Propagation-Based Attribution: Relevance

Propagation

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 ATTENDANCE 1 Talk APPLICABLE
---
--- NOT
Propagation-Based Attribution: Relevance
40 3 PPT APPLICABLE
Propagation
---
--- NOT
5 SUMMARY 2 Talk APPLICABLE
---

SESSION NUMBER : 28

Session Outcome: 1 Students shall analyze model predictions by using propagation-based saliency
methods to visualize feature contributions.

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
40 Propagation-Based Attribution: Saliency Methods 3 PPT APPLICABLE
---
--- NOT

[Link] Page 18 of 29
 26/01/26, 7:18 PM

5 Summary 2 Talk APPLICABLE

---

SESSION NUMBER : 29

Session Outcome: 1 Student shall understand how concept-based methods like Net2Vec, T-CAV, and
ACE explain models using human-understandable concepts.

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
Concept-Based Explainability: Net2Vec, T-CAV,
25 4 PPT APPLICABLE
ACE
---
ALM on Concept-Based Explainability: Net2Vec, T- Group
20 4 Talk
CAV, ACE Discussion

SESSION NUMBER : 30

Session Outcome: 1 Student shall understand how interpretable architectures like CBMs and their
variants provide transparent, concept-driven model reasoning.

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
40 Interpretable Architectures: CBMs and Variants 4 PPT APPLICABLE
---
--- NOT
5 Summary 1 Talk APPLICABLE
---

SESSION NUMBER : 31

Session Outcome: 1 Students shall develop inherently transparent models through interpretable
architectures

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
[Link] Page 19 of 29
 26/01/26, 7:18 PM

---
--- NOT
20 Concept Whitening 4 PPT APPLICABLE
---
--- NOT
20 SENNs 4 PPT APPLICABLE
---
--- NOT
5 Summary 2 Talk APPLICABLE
---

SESSION NUMBER : 32

Session Outcome: 1 Student shall understand how neurosymbolic methods like DeepProbLog and neural
reasoners enable interpretable reasoning.

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
20 Neurosymbolic Methods: DeepProbLog 4 PPT APPLICABLE
---
--- NOT
20 Neural Reasoners 4 PPT APPLICABLE
---
--- NOT
5 Summary 1 Talk APPLICABLE
---

SESSION NUMBER : 33

Session Outcome: 1 STUDENT ABLE TO UNDERSTAND Sample-Based Explanations: Influence

Functions

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 ATTENDANCE 1 Talk APPLICABLE
---
--- NOT
40 Sample-Based Explanations: Influence Functions 3 PPT APPLICABLE
---
--- NOT

[Link] Page 20 of 29
 26/01/26, 7:18 PM

5 SUMMARY 2 Talk APPLICABLE

---

SESSION NUMBER : 34

Session Outcome: 1 student able to apply different sample based explanations in EAI

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
20 Incorporating Explainability into AI Lifecycle 3 PPT APPLICABLE
---
Leading
20 Evaluate the post deployment phase of AI Lifecycle 3 LTC
question
--- NOT
5 Summary 2 Talk APPLICABLE
---

SESSION NUMBER : 35

Session Outcome: 1 STUDENT ABLE TO UNDERSTAND Counterfactual Explanations (conceptual &

causal link)

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 ATTENDANCE 1 Talk APPLICABLE
---
--- NOT
Counterfactual Explanations (conceptual & causal
40 2 PPT APPLICABLE
link)
---
--- NOT
5 SUMMARY 2 Talk APPLICABLE
---

SESSION NUMBER : 36

Session Outcome: 1 Student shall understand how to build explainable AI systems through practical case
studies and projects.

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods

[Link] Page 21 of 29
 26/01/26, 7:18 PM

--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
Case Studies and Projects on Building Explainable
40 4 PPT APPLICABLE
AI Systems
---
--- NOT
5 Summary 1 Talk APPLICABLE
---

SESSION NUMBER : 37

Session Outcome: 1 student able to evaluate Explanation Quality Evaluations

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
20 Incorporating Explainability into AI Lifecycle 5 LTC APPLICABLE
---
Statement-
20 Evaluate the post deployment phase of AI Lifecycle 5 Talk Opinion-
Summary
--- NOT
20 Summary 2 Talk APPLICABLE
---

SESSION NUMBER : 38

Session Outcome: 1 student able to evaluate different metrics of EAI systems

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
20 Incorporating Explainability into AI Lifecycle 5 LTC APPLICABLE
---
--- NOT
20 Evaluate the post deployment phase of AI Lifecycle 5 LTC APPLICABLE
---
--- NOT

[Link] Page 22 of 29
 26/01/26, 7:18 PM

5 Summary 2 Talk APPLICABLE

---

SESSION NUMBER : 39

Session Outcome: 1 Students shall assess the trade-offs between model accuracy and interpretability to
support balanced model selection

Teaching- Active
Time(min) Topic BTL Learning Learning
Methods Methods
--- NOT
5 Attendance 1 Talk APPLICABLE
---
--- NOT
40 Challenges: Accuracy vs Interpretability 4 PPT APPLICABLE
---
--- NOT
5 Summary 2 Talk APPLICABLE
---

Tutorial Course DELIVERY Plan:

NO Delivery Plan Exists

Tutorial Session wise Teaching – Learning Plan

No Session Plans Exists

Practical Course DELIVERY Plan:

NO Delivery Plan Exists

Practical Session wise Teaching – Learning Plan

No Session Plans Exists

Skilling Course DELIVERY Plan:

NO Delivery Plan Exists

Skilling Session wise Teaching – Learning Plan

No Session Plans Exists

WEEKLY HOMEWORK ASSIGNMENTS/ PROBLEM SETS/OPEN ENDEDED PROBLEM-SOLVING EXERCISES

[Link] Page 23 of 29
 26/01/26, 7:18 PM

etc:

Assignment Assignment
Week Topic Details co
Type No
Explain about the
Weekly
different case study
10 Homework 3 Case Studies. CO1
findings through various
Assignments
explainability methods.
Find some hands-on mini
projects, and prepare a
Weekly
study-based report on how
20 Homework 6 Local & Global Methods CO2
explainability is applied
Assignments
using local and global
methods.
Prepare a focused and
well-structured list
explaining the key
components involved in
Monitoring and
Maintaining Explanations
Weekly in Explainable AI (XAI).
Monitoring and Maintaining
24 Homework 7 Your list should highlight CO3
Explanations
Assignments the essential concepts,
methods, and practices
required to ensure that
explanations remain
accurate, stable, and
reliable throughout the
lifecycle of an AI model.
Concept Bottleneck
Models (CBMs) are
interpretable architectures
that require a model to
make predictions through
an intermediate layer of
human-understandable
concepts. Explain the
working mechanism of
Weekly CBMs and their major
Interpretable Architectures:
30 Homework 9 variants, including how CO3
CBMs and Variants
Assignments concepts are defined,
learned, and used for
prediction. Using suitable
examples, analyze the
advantages, limitations,
and practical challenges
of deploying CBMs in
real-world AI systems
where interpretability is

[Link] Page 24 of 29
 26/01/26, 7:18 PM

critical.
Explain the challenges we
Weekly Stability, Robustness, Usability encounter on accuracy vs
38 Homework 12 Metrics interpretability in CO4
Assignments explaining AI models.

Write a one minute paper

describing the
Importance of Explainability in
4 Problem Sets 1 "Importance of CO1
AI Systems
Explainability in AI
Systems".
Form pairs and think on
how AI is applied to
Open Ended
Regulatory Touchpoints: GDPR, analyze the Finance sector
Problem
9 2 Auditability, Case study on (Share market, Stock CO1
Solving
Finance AI exchange, Financial
Exercises
projections etc.) with the
regulatory touch points.
Sample-based explanation
methods (such as
influence functions,
prototypes, and
exemplars) justify model
predictions by referencing
specific training samples.
Open Ended
How do these methods
Problem Sample-Based Explanations:
34 10 determine which samples CO4
Solving ProtoPNets
most strongly influenced a
Exercises
prediction, and in what
ways do such sample-
based explanations help—
or fail—to help—users
understand model
behavior in real-world
tasks?

COURSE TIME TABLE:

Hour 1 2 3 4 5 6 7 8 9
Day Component
Theory --- --- --- --- --- --- V-S51,V-S52,V-S53 --- ---
Tutorial --- --- --- --- --- --- -- --- ---
Mon
Lab --- --- --- --- --- --- -- --- ---
Skilling --- --- --- --- --- --- -- --- ---
Theory --- --- V-S51,V-S52,V-S53 V-S51,V-S52,V-S53 --- --- --- --- ---

[Link] Page 25 of 29