- Handout 0 -- About the course
-
Handout 1 -- Introduction
- Handout 2 -- Unconstrained
optimization
- Handout 3 --
Averaging classifiers -- The 3B's: Bagging, Boosting and Bayesian
Learning
For
additional reading on boosting see the section on boosting on this resources page
especially A
Boosting Approach to Machine Learning by Rob Schapire
and
''Convexity, classification, and risk bounds'' by Peter
L. Bartlett, Michael I. Jordan, Jon D. McAuliffe, in Journal of the
American Statistical Association 101(473), March 2006: 138-156.
For a set of experiments highlighting unanswered questions about
boosting see "Boosting and
the exponential loss" by Abraham Wyner. A journal version of this
paper will be available shortly. - Paul Tseng's 1/29/08
Optimization Seminar slides can be found here
- Divergence measures and message-passing by Tom Minka (paper and slides)
Other resources for approximate inference in grpahical models here
- Handout 5 -- Variational bounds for graphical models
- Handout 6 -- Maximum entropy models
- Handout 7 -- Support vector machines
Pegasos: Primal Estimated sub-GrAdient SOlver for SVM
by Shai Shalev-Shwartz, Yoram Singer, Nathan Srebro
24th International Conference on Machine Learning (ICML), June 2007.
|