CSE 250A. Principles of Artificial Intelligence:
|
Probabilistic methods for reasoning and decision-making under uncertainty. Topics include: inference and learning in directed probabilistic graphical models; prediction and planning in Markov decision processes; applications to computer vision, robotics, speech recognition, natural language processing, and information retrieval.
The course is aimed broadly at advanced undergraduates and beginning graduate students in mathematics, science, and engineering. Prerequisites are elementary probability, multivariable calculus, linear algebra, and basic programming ability in some high-level language such as C, Java, or Matlab. Programming assignments are completed in the language of the student's choice.
CSE 250a covers largely the same topics as CSE 150a, but at a faster pace and more advanced mathematical level. The homework assignments and exams in CSE 250A are also longer and more challenging. In general you should not take CSE 250a if you have already taken CSE 150a.
| Thu Oct 01 | Administrivia and course overview. | |
| Tue Oct 05 | Modeling uncertainty, review of probability, explaining away. | HW 1 out. |
| Thu Oct 08 | Belief networks: from probabilities to graphs. | |
| Tue Oct 13 | Representing conditional probability tables. Conditional independence and d-separation. | HW 1 due. HW 2 out. |
| Thu Oct 15 | Probabilistic inference in polytrees. | |
| Tue Oct 20 | More algorithms for inference: node clustering, cutset conditioning, likelihood weighting. | HW 2 due. HW 3 out. |
| Thu Oct 22 | Markov Chain Monte Carlo algorithms for inference. Learning from complete data. | |
| Tue Oct 27 | Maximum likelihood estimation. Markov models of language. Naive Bayes models of text. | HW 3 due. HW 4 out. |
| Thu Oct 29 | Linear regression and least squares. Detour on numerical optimization. | |
| Tue Nov 03 | Logistic regression, gradient descent, Newton's method. Learning from incomplete data. | HW 4 due. HW 5 out. |
| Thu Nov 05 | EM algorithm for discrete belief networks: derivation and proof of convergence. | |
| Tue Nov 10 | EM algorithms for word clustering and linear interpolation. | HW 5 due. HW 6 out. |
| Thu Nov 12 | EM algorithms for noisy-OR and matrix completion. Discrete hidden Markov models. | |
| Tue Nov 17 | Computing likelihoods and Viterbi paths in hidden Markov models. | HW 6 due. HW 7 out. |
| Thu Nov 19 | Forward-backward algorithm in HMMs. Gaussian mixture models. | |
| Tue Nov 24 | Linear dynamical systems. Reinforcement learning and Markov decision processes. | HW 7 due. HW 8 out. |
| Thu Nov 26 | Thanksgiving holiday. | |
| Tue Dec 01 | State and action value functions, Bellman equations, policy evaluation, greedy policies. | HW 8 due. HW 9 out. |
| Thu Dec 03 | Policy improvement and policy iteration. Value iteration. Algorithm demos. |
|
| Tue Dec 08 | Convergence of value iteration. Model-free algorithms. Temporal difference prediction. | |
| Thu Dec 10 | Q-learning, RL in large state spaces. Bonus topics. Course wrap-up. |
HW 9 due |
| Sat Dec 12 | Take-home final exam (released). |