Course Description and Prerequisities

COMP 540 is about learning models from data. The course is designed to give you a foundational understanding of modern algorithms in learning and data mining, as well as hands-on experience with its applications in science and engineering.

The course assumes familiarity with discrete mathematics (COMP 280) and a facility with algorithms as well as with programming (COMP 210 and COMP 212). COMP 314 and COMP 440 are highly recommended. COMP 540 can be taken as grounding for future research in machine learning or to gain familiarity with machine learning methods for application in other fields.

Texts and Reading Material

There is no course textbook. Recommended texts on statistical machine learning include:

The elements of statistical learning T. Hastie, R. Tibshirani and J. Friedman, Springer, 2001.
Neural networks for pattern recognition Christopher Bishop, Oxford University Press, 1996.
Pattern Recognition 2nd edition, R. O. Duda and P. E. Hart and D. G. Stork, Wiley-Interscience, 2001.
Machine Learning Tom Mitchell, McGraw Hill, 1997.

Here are two specialized books on machine learning covering two specific classes of methods: learning in graphical models and support vector machines.

Learning in graphical models edited by M. Jordan, MIT Press, 1998.
Learning with kernels B. Schölkopf and A. Smola, MIT Press, 2002.

Last modified: 5 January 2008 by

Devika Subramanian
devika@rice.edu