Advanced Topics in Statistical Machine Learning COMP8650

Learning outcomes

More information may be available for enrolled students on the course website on Wattle

More information may be available for enrolled students on the course website at http://cecs.anu.edu.au/courses/info/comp8650/overview

At the end of the course students should be able to:

1. write down definitions of key concepts in convex analysis, including convexity of sets and functions, subgradients, and the convex dual
2. derive basic results about convex functions such as Jensen’s inequality
3. understand how Bregman divergences are constructed from convex functions and derive some of their properties
4. write down a formal optimization problem from a high-level description and determine whether the problem is convex
5. recognize standard convex optimization problems such as linear programs and quadratic programs
6. derive the standard (dual) quadratic program for support vector machines and understand the extension to max-margin methods for structured prediction
7. implement and analyse gradient descent algorithms such as stochastic gradient descent and mirror descent