STAT 339 - Probabilistic Modeling and Machine Learning

Semester Offered: Second Semester
4NS, QFR
Credits: 4 credits
Attribute: 4NS, QFR

An overview of statistical models and algorithms used in machine learning for classification, prediction, clustering, hidden variable modeling, and sequence learning. Fundamentals of probability, Bayesian inference and decision theory, model selection, and stochastic optimization. Modeling approaches include directed and undirected graphs, parametric, nonparametric and semi-parametric mixture models, Hidden Markov Models, and selected non-probabilistic techniques such as Support Vector Machines and Neural Networks. Emphasis throughout is on probabilistic reasoning from data. Applications selected from a variety of domains, based on student interest.

Enrollment Limit: 28
Instructor: C. Dawson

Prerequisites & Notes: MATH 231, CSCI 150 and at least some additional experience with linear algebra and/or probability would be helpful.