Introduction to Machine Learning (Aug 2022)

This contains the details of the tutorial sessions (1.5 hours each) conducted during this course. Coding tutorials (in python) related to the topics covered in class were also conducted in most sessions.

Tutorial 1: Introduction to ML: bias-variance tradeoff, contingency table, naive classifier coding (only numpy)

Tutorial 2: Supervised Learning: Entropy, Decision Trees, Linear Regression as projection onto column space, coding linear regression (only numpy)

Tutorial 3: Dimensionality Reduction: KNN, projection (Linear Algebra), similarity measure, PCA: eigen vector view and coding (only numpy)

Tutorial 4: Bayes Learning: Bayes Rule, Discriminant Functions, MLE, MAP, and Bayes Networks: problem solving, equivalence of Regularization and prior in MAP

Tutorial 5: Logistic Regression, SVMs: Gradient Descent (Logistic Regression), SVM (soft and hard), and problem-solving related to SVMs. Tutorial on how to code SVM.

Tutorial 6: Artificial Neural Networks: Multilayer perceptron, Backpropagation (vectorized implementation for MLPs covered), computational graphs, coding MLP from scratch and using TensorFlow, and introduction to CNNs

Tutorial 7: Computational Learning Theory: PAC model numericals, shattering and VC-dimension, Ensemble Learning, Recap of Bagging and Boosting, and demonstration of the AdaBoost algorithm

Tutorial 8: Unsupervised Learning: Langrangian Multiplier, solving a constrained optimization problem, Soft clustering, K-means theory and coding, Agglomerative Clustering

Tutorial 9: Revision: Some good practices in ML. Revision: Metric for evaluation, linear regression, Bayesian Learning, SVMs and PCA

You can get the tutorial materials here, python implementations here, and the video lectures in this youtube link. Certificate