CSE 291. Probabilistic Methods in AI and Machine Learning

Subject

Probabilistic methods for reasoning and decision-making under uncertainty. Inference and learning in probabilistic graphical models; prediction and planning in Markov decision processes; applications to computer vision, robotics, speech recognition, natural language processing, and information retrieval.

Prerequisites

The course is aimed at advanced undergraduates and beginning graduate students in mathematics, science, and engineering. Prerequisites are multivariable calculus, linear algebra, elementary probability, and basic programming ability in some high-level language such as C, Java, or Matlab.

Texts

• S. Russell and P. Norvig, Artificial Intelligence: A Modern Approach, Prentice Hall, 2003.
• R. Sutton and A. Barto, Reinforcement Learning: An Introduction, MIT Press, 1998.
• C. Bishop, Pattern Recognition and Machine Learning, Springer, 2006.
• T. Hastie, R. Tibshirani, and J. Friedman, Elements of Statistical Learning, Springer-Verlag, 2001.

Instructors

• Professor: Lawrence Saul
• Grader: Chih-Chieh Cheng

Meetings

• Lectures: Tue/Thu 11-12:30, WLH 2208.
• Sections: Thu 3-4, EBU3B Room 4217

Grading

• homework assignments (50%)
• midterm exam (20%)
• final exam (30%)

Assignments

• Assignment 1 - due Fri Jan 19
• Assignment 2 - due Fri Jan 26
• Assignment 3 - due Fri Feb 2
• Assignment 4 - due Fri Feb 9 [data]
• Assignment 5 - due Fri Feb 23 [data]
• Assignment 6 - due Fri Mar 02 [data]
• Assignment 7 - due Tue Mar 13 [data]

Syllabus

• Tue Jan 09 - Administrivia and course overview.
• Thu Jan 11 - Modeling uncertainty, review of probability, explaining away.
• Tue Jan 16 - Belief networks: from probabilities to graphs.
• Thu Jan 18 - Conditional independence, d-separation, polytrees.
• Tue Jan 23 - Algorithms for exact and approximate inference.
• Thu Jan 25 - Learning via maximum likelihood estimation.
• Tue Jan 30 - Markov models of language, naive Bayes for text classification.
• Thu Feb 01 - Linear and logistic regression.
• Tue Feb 06 - Latent variable models, EM algorithm.
• Thu Feb 08 - EM algorithms for statistical language modeling.
• Tue Feb 13 - Hidden Markov models (HMMs), automatic speech recognition.
• Thu Feb 15 - Midterm exam.
• Tue Feb 20 - Viterbi and forward-backward algorithms in HMMs.
• Thu Feb 22 - Multivariate Gaussian distribution, mixture models, Kalman filtering.
• Tue Feb 27 - Reinforcement learning, Markov decision processes (MDPs).
• Thu Mar 01 - Policy evaluation, policy improvment, policy iteration.
• Tue Mar 06 - Value iteration, stochastic approximation theory.
• Thu Mar 08 - Temporal difference learning, Q-learning, applications and extensions.
• Tue Mar 13 - Course wrap-up and evaluations.
• Thu Mar 15 - Final exam.