CMPSCI 689

Machine Learning

Fall 2011


Tentative Course Schedule

(B refers to Bishop's text; CB refers to Casella and Berger; DHS refers to Duda, Hart, and Stork; HTF refers to the book by Hastie, Tibshirani, and Friedman; M refers to Mackay's book; W refers to Weber).

Lecture Date Topic Textbook Reading Background Reading Homework Assigned Homework Due
Lecture 1 Wed Sept 7th Overview of Machine Learning Sec 1.1 (B), Chapter 2 (HTF). Read this article on a computational theory of learning, due to Les Valiant, winner of the 2011 Turing Award.
Lecture 2 Mon Sept 12th Statistical Models Sec 1.2, Sec 2.1-2.3 (B), Sec. 2.1-2.4, 23.1-23.3 (M). Read this article on the $1 million Netflix datamining contest
Lecture 3 Wed Sept 14th Regression, Regularization and Model Selection Sec 1.1, 1.3, 3.1 (B), Sec 3.1-3.4 (HTF). Read this article on least angle regression (Sections 1-3). Homework 1
Lecture 4 Mon Sept 19th Linear classification; Bayesian Decision Theory Sec 1.5, Sec 4.1-4.2 (B), Ch. 2.1-2.6 (DHS). Read this article on self taught learning .
Lecture 5 Wed Sept 21st Causal and Graphical Models Chapter 8 (B) . Read this review article on causal modeling .
Lecture 6 Mon Sept 26th Principles of Statistical Data Reduction Chapter 6 (CB). Read this review article on conditional independence in statistics . HW 1
Lecture 7 Wed Sep 28th Maximum Likelihood and Bayesian Estimation Sec. 2.3-2.4 (B), Ch. 2, 5 (W), Ch. 22 (M). Read this review article on the optimality of the naive Bayes classifier . Homework 2
Lecture 8 Mon Oct 3th Bias Variance Analysis; Cramer-Rao Theorem Sec. 3.2 (B), Ch. 3 (W). Read this article on probabilistic canonical correlational analysis .
Lecture 9 Wed Oct 5th Expectation Maximization Algorithm; Mixture Models Ch. 9 (B), Ch. 20 (M). Read this article on scientific topic modeling .
Lecture 10 Tue Oct 11th (Mon schedule!) Project Proposals Read this article on 3D Map Learning on mobile robots using EM . HW 2
Lecture 11 Wed Oct 12th Time-series data and Hidden Markov Models Ch. 13 (B). Read this article on a variational algorithm for learning abstract hidden Markov models . Homework 3
Lecture 12 Mon Oct 17th Learning to Control Dynamical Systems Read this Survey article .
Lecture 13 Wed Oct 19th Logistic Regression; Generalized Linear Models Sec. 4.3 (B), Sec. 4.4 (HTF) HW 3
Lecture 14 Mon Oct 24th Maximum Entropy Framework Midterm Exam Sample midterm exam
Lecture 16 Wed Oct. 26th Instance-based Learning; Kernel Methods Ch. 6 (B) Read this article on maximum entropy methods .
Lecture 17 Mon Oct. 31st Support Vector Machines Ch. 7 (B), Ch. 12 (HTF) Read this article on string kernels .
Lecture 18 Wed Nov 2nd Reproducing Kernel Hilbert Spaces Homework 4
Lecture 19 Mon Nov 7th Computational Learning Theory Ch. 7 (M) Read this classic article by Gold on language identification in the limit , and Valiant's A Theory of the Learnable .
Lecture 20 Mon Nov 9th Dimensionality Reduction Ch 12 (B), Sec. 14.4-14.7 (HTF), Section 3.8 (DHS)
Lecture 21 Mon Nov 14th Manifold Learning Read this overview of spectral methods for dimensionality reduction
Lecture 22 Mon Nov 16th Learning Representation and Control in MDPs Read this article HW 4
Lecture 23 Mon Nov 21st Project Status Report
Lecture 24 Mon Nov 28th Future Directions in ML
Wed Nov. 30th Future Directions in ML
Mon Dec 5th Project Presentations
Wed Dec 7th Project Presentations
Fri Dec 9th Final Project Reports and Term Papers due