This course covers machine learning (ML) via a probabilistic model-based approach informed by Bayesian inference. This provides a structure for developing techniques and systems that can address a wide range of problems relating to: inferring data processes, prediction, generation, discovering latent structures and decision-making. Fundamentals of probability, statistics, graphical modeling, information theory and optimization will also be covered in the early part of the course. Students will gain a deep understanding of the probabilistic approach to ML through lectures, problem sets, two midterms and a final project. Theoretical considerations will be balanced by applications to engineering and scientific problems.
The main text for the course is Kevin’s Murphy’s Probabilistic Machine Learning: Advanced Topics