BBF 304E - Learning From Data

Course Objectives

Introduce students to major data analytics and machine learning methods and underlying theories
Learning to apply available tools to solve classification, clustering and regression problems
Learning to avoid major pitfalls such as overfitting, confusing correlation and causality whilile using machine learning tools
Learning the assessment and comparison of performance of machine learning methods

Course Description

Introduction to Machine Learning, major applications Mathematical background, marginal and conditional Probability, Bayes theorem, Bayesian decision theory Density estimation, Maximum Likelihood estimate, Bayesian Learning, Naïve Bayes Linear regression Bias-variance dilemma, regularization, ridge regression and lasso Linear classifiers Artificial neural networks, perceptron and multilayer perceptron Assessment and comparison of classifier performance Feature selection and extraction Large margin classifiers, support vector machines, kernel methods Decision trees and random forest Unsupervised learning, clustering Deep learning and big data