Department of Information Technology

Statistical Machine Learning (SML) PhD course (9+3hp)

Spring 2016

Lectures

Each lecture comes with a list of recommended problems to be solved as shown in the table below. If there are no letters in front of the numbers, they refer to problems in the book by Bishop. If the letters HTF appears in front of the number that means that exercise is to be found in the book by Hastie, Tibshirani and Friedman.

Note that the slides provided below only covers a small part of the lectures, the blackboard is used quite extensively.

The schedule is available via TimeEdit by clicking here.

Nr. Contents Chapter Pres. Problems
1. Introduction 1-2, notes le1 2.13, 2.29, 2.32, 2.34, 2.40, 2.44, 2.47.
2. Linear regression 3, HTF:3 le2 1.25, 1.26, 3.8, 3.9, 3.12, 3.13.
3. Linear classification 4 le3 4.5, 4.19, 4.25, HTF:2.8.
4. (Deep) neural networks and kernel introduction 5 le4 5.4, 5.16, HTF:11.5.
5. Kernel methods, Gaussian processes 6-7 le5 6.3, probl. m-file
6. EM and clustering 9, notes le6 9.8, 9.9, 9.11, 12.24 (also in Matlab, see lecture 1).
7. Approximate inference 10, notes, code le7 10.4, 10.7, 10.26, 10.38.
8. Graphical models 8 le8 14.6, 14.7, 8.1, 8.3, 8.4, 8.7.
9. Graphical models and message passing 8, code le9 8.10, 8.11, 8.19, 8.23, 8.27.
10. MCMC and sampling methods 11, code le10 probl m-file
11. Bayesian nonparametric models P1, P2, P3, code le11
Updated  2016-03-22 17:23:42 by Fredrik Lindsten.