Fall 2003
Fall 2002
Fall 2001

Lectures

Date	Lecture	Notes etc
Wed 9/8	Lecture 1: introduction pdf slides, 6 per page
Mon 9/13	Lecture 2: linear regression, estimation, generalization pdf slides, 6 per page	(Jordan: ch 6-6.3)
Wed 9/15	Lecture 3: additive regression, over-fitting, cross-validation, statistical view pdf slides, 6 per page
Mon 9/20	Lecture 4: statistical regression, uncertainty, active learning pdf slides, 6 per page
Wed 9/22	Lecture 5: from regression to classification, decision theory, logistic regression pdf slides, 6 per page
Mon 9/27	Lecture 6: logistic regression, regularization, discriminative classification pdf slides, 6 per page
Wed 9/29	Lecture 7: support vector machines, kernels pdf slides, 6 per page	Notes on Lagrange multipliers (postscript) Optional reading: Burges (postscript)
Mon 10/4	Lecture 8: kernel methods, kernels pdf slides, 6 per page
Wed 10/6	Lecture 9: feature selection, combination of methods, forward-fitting pdf slides, 6 per page
Wed 10/13	MIDTERM: in class
Mon 10/18	Lecture 10: boosting pdf slides, 6 per page	Optional reading: Schapire et al (postscript) Friedman et al (postscript)
Wed 10/20	Lecture 11: complexity, VC-dimension, learning pdf slides, 6 per page
Mon 10/25	Lecture 12: VC-bounds, structural risk minimization, compression and model selection pdf slides, 6 per page
Wed 10/27	Lecture 13: Minimum description length principle; structure, mixtures, and the EM-algorithm pdf slides, 6 per page
Mon 11/1	Lecture 14: The EM-algorithm and Gaussian mixtures; convergence, regularization, and classification pdf slides, 6 per page
Wed 11/3	Lecture 15: Mixture classifiers, mixtures of experts; non-parametric mixtures; clustering pdf slides, 6 per page
Mon 11/8	Lecture 16: clustering; k-means and spectral. pdf slides, 6 per page
Wed 11/10	Lecture 17: clustering; semi-supervised and model based pdf slides, 6 per page
Mon 11/15	Lecture 18: Hidden Markov Models pdf slides, 6 per page
Wed 11/17	Lecture 19: HMMs, EM, viterbi pdf slides, 6 per page
Mon 11/22	Lecture 20: Graphical models (Bayesian networks) pdf slides, 6 per page
Wed 11/24	Lecture 21: Undirected graphical models, medical diagnosis, inference and messages pdf slides, 6 per page
Mon 11/29	Lecture 22: Exact probabilistic inference, message passing pdf slides, 6 per page
Wed 12/1	Lecture 23: Exact inference and junction trees; learning Bayesian networks pdf slides, 6 per page	Projects due Fri Dec 3!
Mon 12/6	Lecture 24: Learning Bayesian networks; review for the final pdf slides, 6 per page
Wed 12/8	FINAL EXAM: in class