Statistical Learning
Modeling, Prediction and Computing
STAT 535 Autumn Quarter 2011


Course Description


Books and other resources

Class mailing list



Handouts/Course notes

UW Statistics


What will the course be about?

  • The theme of STAT 535 is unsupervised learning with emphasis on clustering, i.e finding groups in data, and on graphical probability models (aka belief networks). Both are outstanding examples of statistics and algorithms at work together.
Who is this class for?
This class is the first in the Statistics PhD Learning sequence, but it is regularly attended by other students with an interest in machine learning, graphical models and the connection of statistics to algorithms and optimization.


  • A course in probability, including basic notions of multivariate analysis (conditional probability, marginals).
  • Algorithms and data structure at a basic level (arrays, lists, sets, O( ) notation).
  • Knowledge of a computer programming language (like C, C++, Java, Matlab, R, Splus) For Statistics students, taking STAT 534 is a good way to get up to speed in algorithms and programming.

Instructor: Marina Meila   mmp at stat dot washington dot edu

Lectures: Tuesdays & Thursdays 11:30 - 12:50 in GUG 204

Office hours: Monday 2-3pm in PDL B-321

Course home page: (this page)

Class mailing list: stat535a_au11 at UW