Computer Science 246/446

Mathematical Foundations of Artificial Intelligence

Spring 2007

Instructor: Dan Gildea office hours 1-2pm Wed
TA: Shaojun Zhao office hours 1-2pm Wed
Location: Tu/Th 11:05am-12:20pm, CSB 601

Homeworks

Required text: Christopher M. Bishop, Pattern Recognition and Machine Learning.

The following are useful references in addition to the reading material assigned for each class:

Readings not taken from Bishop are available in the box in the CS mailroom.

Syllabus

Onwe will coverwhich means that after class you will understandif before class you have read
1/18 Probability Theory independence, bayes rule bishop 1.2
1/23 Information Theory entropy, kl-distance, coding bishop 1.6
1/25 Probabilistic Inference priors on continuous variables bishop 2.2
1/30 Probabilistic Inference priors: bayesian reasoning, MAP bishop 3.4
2/1 Perceptron, Backpropagation the chain rule bishop 5.1, 5.2, 5.3
2/6 Support Vectors the kernel trick bishop 6.1, 6.2
2/8 Support Vectors the wolfe dual bishop 7.1
2/13 Probabilistic Inference polytree bishop 8.4
2/15 Expectation Maximization latent variable clustering bishop 9
2/20 Sampling Markov Chain Monte Carlo bishop 11.2
2/22 Sampling Gibbs Sampling bishop 11.3
2/27 Eigenvectors least squares, PCA bishop 12.1
3/1 Learning Theory probably approximately correct kearns&vazirani 1
3/6 Learning Theory VC dimension kearns&vazirani 2, 3
3/8 Independent Component Analysis source separation bishop 12.4.1
3/20 Review
3/22 Midterm
3/27 Midterm Solutions
3/29 Logistic Regression maximum entropy bishop 4.3
4/3 Optimization conjugate gradient bishop (1995) 274-282
4/5 Hidden Markov Models forward-backward bishop 13.2
4/10 Particle Filters slam Thrun et al. ch 4
4/12 Reinforcement Learning q-learning ballard 11
4/17 Reinforcement Learning partial observability ballard 11
4/19 Games nash equilibrium morris 115-131
4/24 Games learning to co-operate
4/26 Something Fun
5/1 Review come to class with questions!
Final Exam: Friday May 11, 8:30am.

Grading


gildea @ cs rochester edu
April 4, 2007