CS460303 Machine Learning L-T-P-Cr: 3-0-2-4
Pre-requisites: Fundamental knowledge of AI, linear algebra, probability & statistics, and algorithms
Objectives/Overview:
● To learn and understand machine learning and its types.
● To familiarize with Python programming language.
● To learn common unsupervised machine learning approaches.
● To learn common supervised machine learning approaches.
● To understand single layer and multi-layer neural network models.
● To learn to apply common machine learning techniques for solving real world problems.
Course Outcomes:
At the end of the course, a student should:
Sl. Outcome Mapping to POs
No.
1. Have knowledge of fundamental aspects of machine learning PO2
approach.
2. Be able to choose appropriate regression technique for PO1, PO2,PO3
modelling real life problems.
3. Have algorithmic knowledge of common clustering techniques. PO3
4. Understand the purpose of dimensionality reduction and PO2,PO3
unsupervised PCA technique
5. Have knowledge of simple classification techniques and able to PO1, PO2
classify simple datasets using them.
6. Conceptual understand the working of SVM and “Kernel Trick” PO1, PO2
7. Have algorithmic knowledge of supervised LDA dimensionality PO1, PO3
reduction technique.
8. Understand of the purpose of ensemble learning. PO1, PO2
9. Conceptually understand reinforcement learning and be able to PO2
iterate on simple problem datasets
10. Algorithmic knowledge of working of common neural network PO1, PO2
models and be able to iterate on simple problem datasets.
11. Be able to derive back-propagation of error for multi-layer PO1, PO3, PO12
neural networks.
UNIT I: Introduction Lectures:6
Introduction to machine learning, types of learning, common aspects of machine learning
approach: model, parameters, Bias-Variance. Test, train and validation datasets, error
function. Curse of dimensionality. Predictive analysis using regression.
UNIT II: Python Basics Lectures: 6
Fundamental data types in Python, looping and decision making constructs, functions,
classes, file handling, database access, output formatting, classes, modules, statistics module,
numpy, introduction to popular machine learning libraries TensorFlow & Keras.
UNIT III: Neural Networks Lectures: 12
Introduction, perceptron model, gradient descent algorithm, learning mechanisms and
activation functions, multi-layer feed forward, back-propagation, introduction to feed-back
networks.
UNIT IV: Supervised Learning Lectures: 6
Types of supervised learning, classification: KNN, Bayes, Decision Tree. SVM: soft and
hard margin, kernel trick. LDA – supervised dimensionality reduction technique.
UNIT V: Unsupervised Learning Lectures: 6
Types of unsupervised learning, clustering, K-means, GMM & EM Algorithm. Eigen values
and Eigen vectors, PCA – unsupervised dimensionality reduction technique.
UNIT V: Ensemble and Reinforcement Learning Lectures: 6
Introduction to reinforcement learning, bagging, Random Forest and Boosting. Q-learning
and SARSA algorithms
Text/Reference Books
1) Machine Learning. Tom Mitchell, McGraw-Hill.
2) Machine learning: an algorithmic perspective. Marsland, Stephen. Chapman and Hall/CRC,
2011.
3) Introduction to artificial neural systems. Zurada, Jacek M. Vol. 8. St. Paul: West publishing
company, 1992.
4) A Tutorial on Support Vector Machines for Pattern Recognition. Christopher Burges, Data
Mining and Knowledge Discovery, 1998.