Information about CS 830 (Machine Learning), Fall 2011. This page will be updated during the term.


(A) Formal learning frameworks
A.1 Probably approximately correct learning
A.2 Learning from queries
A.3 Learning from teachers

(B) Applied models of learning
B.1 Tree-based models (e.g., decision trees)
B.2 Maximum margin models (e.g., support vector machines)
B.3 Clustering models (e.g., k-Means, spectral clustering)
B.4 Combinations of models (e.g., boosting)


15% assignments
10% project proposal (due Sep 30)
20% literature summary related to the research project (due Oct 31)
15% oral presentation of research results (near the end of term)
40% project report (due near the end of term)

Projects can be experimental, theoretical, or both (some choices will be given, but students may have to come up with their own project idea).
Students are supposed to choose a research topic on their own.

Some suggestions for possible choices are available here.
The course does not follow a textbook. However, useful pointers can be found here.