EC ENGR 246
Foundations of Statistical Machine Learning
Description: Lecture, four hours; discussion, one hour; outside study, seven hours. Enforced requisites: course 131A, Mathematics 33A. Introduction to foundations of statistical machine learning. Overview of several widely used learning algorithms including logistic and linear regression, kernel methods and support vector machine (SVM), ensemble learning methods, decisions trees and nearest neighbor classifiers. Connections to information theory through probably approximately correct (PAC) learning, stability, bias-complexity trade-off, structural risk minimization, minimum description length (MDL), and universal learning. Introduction to representation learning with topics including unsupervised learning, clustering, (non-linear) dimensionality reduction, sketching, parametric distribution estimation including Gaussian mixtures, expectation maximization, non-parametric distribution estimation, property testing and neural networks focused on distribution sampling (variational autoencoders ÝVAEs¨, generative adversarial networks ÝGANs¨). Discussion of reinforcement learning. Letter grading.
Units: 4.0
Units: 4.0
Most Helpful Review
Winter 2025 - The material is challenging and requires mathematical maturity. The recorded lectures and reading material provided will help you a lot in learning the material. Learning the material also demands for a lot of self studying and even more so if you want to do well in the final exam and be able to do your assignments. The Professor also makes the lectures interesting by sharing anecdotes and developments during his time at Bell Laboratories. You will love the course if you enjoy Probability.
Winter 2025 - The material is challenging and requires mathematical maturity. The recorded lectures and reading material provided will help you a lot in learning the material. Learning the material also demands for a lot of self studying and even more so if you want to do well in the final exam and be able to do your assignments. The Professor also makes the lectures interesting by sharing anecdotes and developments during his time at Bell Laboratories. You will love the course if you enjoy Probability.