M.Sc. (Information Technology) Machine Learning Syllabus - Mumbai University

• What Is Machine Learning? Basic Concepts in Machine Learning - Classification versus Regression, Supervised versus Unsupervised Learning Simple versus Complex Models, Parametric versus Nonparametric Models Overfitting versus Underfitting, Bias–Variance Trade-Off General Principles in Machine Learning Occam’s Razor, No-Free-Lunch Theorem, Law of the Smooth World, Curse of Dimensionality. Advanced Topics in Machine Learning Reinforcement Learning, Meta-Learning, Causal Inference, and Other Advanced topics Mathematical Foundation Probability and Statistics Random Variables and Distributions, Expectation: Mean, Variance, and Moments, Joint, Marginal, and Conditional Distributions, Common Probability Distributions, Transformation of Random Variables Information Theory Information and Entropy, Mutual Information, KL Divergence Mathematical Optimization General Formulation, Optimality Conditions, Numerical Optimization Methods

• Introduction, Decision-based methods: Nonlinear Instance-based methods, Decision Tree algorithm, Use Cases (Reference 2) Feature Extraction Feature Extraction: Concepts,Linear Dimension Reduction, Nonlinear Dimension Reduction (I): Manifold Learning, Nonlinear Dimension Reduction (II): Neural Networks MODULE II: Models in Machine Learning and Applications

• Statistical Learning Theory, Linear Models, Learning Discriminative Models in General, Neural Networks, Ensemble Learning Generative Models Overview of Generative Models, Formulation of Generative Models, Bayesian Decision Theory, Statistical Data Modeling, Density Estimation, Maximum- Likelihood Estimation, Maximum-Likelihood Classifier, Generative Models (in a Nutshell) Generative versus Discriminative Models Unimodal Models Gaussian Models, Multinomial Models, Markov Chain Models, Generalized Linear Models Mixture Models Formulation of Mixture Models, Expectation-Maximization Method, Expectation-Maximization Method, Gaussian Mixture Models, Hidden Markov Models Entangled Models Formulation of Entangled Models, Linear Gaussian Models, Non-Gaussian Models, Deep Generative Models Bayesian Learning Formulation of Bayesian Learning, Conjugate Priors, Approximate Inference, Gaussian Processes Graphical Models Concepts of Graphical Models, Bayesian Networks, Markov Random Fields

• Learning (Reference 2)

Books and References

• 1 Machine Learning Fundamentals: A Hui Jiang Cambridge University First 2021 Concise Introduction Press
• 2 Machine Learning Concepts, T V Geetha, CRC Press Taylor First 2023 Techniques and Applications SSendhilkumar & Francis Group
• 3 Mastering Classification Partha BPB First 2023 Algorithms for Machine Learning Majumdar
• 4 Machine Learning for Absolute Oliver Scatterplot Press Third 2021 Beginners Theobald
• 5 Machine Learning: The Art and Cambridge Science of Algorithms that Make Peter Flach University First 2012 Sense of Data Press Online Instructor slides https://github.com/iNCML/MachineLearningBook Resources