Current course offerings

Please Note: Course profiles marked as not available may still be in development.

Course description

The course focuses on the statistical mathematical understanding of machine learning and data science. Topics to be covered include theoretical frameworks for supervised learning, unsupervised learning, and semi-supervised learning. The use of mixture models for statistical learning and clustering will be covered. Approaches to the assessment of multivariate normality will be initially considered. The Akaike and Bayesian Information criteria for model choice will be derived. The bootstrap will also be considered for model choice and parameter estimation. Methods for the analysis of high-dimensional data will be covered, including the use of principal components and more complex models such as mixtures of factor analyzers. Software for the implementation of the various methods will be covered.

Archived offerings