• LO1: To introduce different machine learning paradigms
• LO2: To provide understanding of machine learning algorithms to be used on a given dataset for regression/classification problems.

• CO1: Ability to conduct data analysis and data visualization
• CO2: apply the complete ML pipeline in real-world dataset – Analyse datasets, decide pre-processing steps, visualize data, apply ML models, and infer the meaning based on different performance metrics.

Role of learning in intelligent behaviour, general structure of a learning system; learning from example; concept learning, Introduction to machine learning and machine learning applications, Supervised learning, linear regression, polynomial regression, logistic regression, multivariate methods, dimensionality reduction, Support Vector Machine, clustering. Neural networks, multilayer perceptron, local models, assessing and comparing ML models. MLOps – introduction to converting ML models from test bench to production (saving, loading, using trained models).

Machine Learning practical exercises:

1. Design and implementation of a Bayes classifier for two-class and multi-class classification
2. Design and implementation of an MLP based Artificial Neural Network Model for classification or regression
3. Design and implementation of a deep learning classifier model using transfer learning
4. Design and implementation of a simple DAG Network for deep learning
5. Design and implementation of clustering algorithms
6. Determining the Bipartiteness of a graph using search algorithms

Recommended Tools: MATLAB, Python

1. Tom. Mitchell, “Machine Learning”, McGraw Hill, 1997.
2. Géron, Aurélien. Hands-on machine learning with Scikit-Learn, Keras, and TensorFlow: Concepts, tools, and techniques to build intelligent systems. O’Reilly Media, 2019.