Announcements: 

Brief description
This course presents several landmark methods of Machine Learning in a unified framework. We aim:

(1) to introduce the ideeas and mathematics that are fundamental to any problem that involves learning from data, and

(2) to teach a selection of important and general methods of machine learning

Hence, we will emphasize equally the techniques themselves and the common underlying concepts. Many of these concepts are rooted in probability and statistics, hence we will often adopt a probabilistic perspective to describe the problem setup and the solution. Then we will analyze how the solution is obtained from an algorithmic perspective.

Lectures: Tuesdays 3:30 - 4:50, Fridays 1:00 - 2:20 in MGH 295
Office hours:TBA

Prerequisites: basic notions of probability (at the level of STAT 390/391/394), basics of algorithms and complexity, multivariate calculus. We will introduce the statistics concepts that we are going to use during the course, as well as notions of information theory.

Format: Two 1.5 hour lectures + discussions weekly. Typically, the instructors will start by presenting each topic, then we will progress towards an open discussion. The reading materials will be course notes plus original research and tutorial papers.

Graded: 3 credits C/NC based on two mini-projects + homework + class participation.

Course home page: http://www.ms.washington.edu/stat592/winter04 (this page)