Schedule

Note: topics and deadlines are subject to change.

Date		Topic	Readings	Due
08/31	Th	Introduction to Rational Agents states, actions, optimality, rationality	Ch 1-2
Module 1: Search
09/05	Tu	Search Algorithms transition model, cost function, search tree	Ch 3.1-3.3
09/07	Th	Uninformed Search 1 breadth-first, depth-first, iterative-deepening	Ch 3.3-3.4
09/12	Tu	Uninformed Search 2 best-first, uniform-cost	Ch 3.3-3.4	HW 1
09/14	Th	Informed Search a-star, heuristic, admissibility, consistency	Ch 3.5-6
09/19	Tu	Adversarial Search zero-sum, mini-max, alpha-beta pruning	Ch 5	HW 2
Module 2: Markov Decision Processes
09/21	Th	Markov Decision Processes policy, reward, transition, utility, discount	Ch 17.1
09/26	Tu	Solving MDPs 1 value, q-value, value iteration, bellman equation	Ch 17.2	HW 3
09/28	Th	Solving MDPs 2 policy iteration	Ch 17.2	Project 1 (Search)
10/03	Tu	Passive Reinforcement Learning bandit problems, temporal-difference learning, q-learning	Ch 22.2	HW 4
10/05	Th	Active Reinforcement Learning exploration vs exploitation, epsilon-greedy, exploration fn	Ch 22.3
Module 3: Probabilistic Inference
10/10	Tu	Representing Uncertainty joint, marginal, and conditional probability	Ch 12.1-12.3	HW 5
10/12	Th	Bayes’ Rule chain rule, product rule, inference, independence	Ch 12.4-12.6	Project 2 (MDPs)
10/17	Tu	Markov Models POMDPs, belief state, markov chain	Ch 17.4, 14.1	HW 6
10/19	Th	No Class October Break
10/24	Tu	Inference in Temporal Models 1 filtering, state estimation, prediction, smoothing	Ch 14.2
10/26	Th	Inference in Temporal Models 2 forward-backward, viterbi algorithm, HMMs	Ch 14.3
10/31	Tu	Particle Filters 1 sampling, kalman filter	Ch 14.4-14.5	HW 7
11/02	Th	Particle Filters 2 dynamic bayesian network, MCMC	Ch 14.4-14.5
Module 4: Classification and Regression
11/07	Tu	Learning from Examples naive bayes, decision tree, bias-variance tradeoff	Ch 12.6, 19.3	HW 8
11/09	Th	Information Theory under/over-fitting, entropy, information gain	Ch 19.2-19.3	Project 3 (Inference)
11/14	Tu	Ensemble Learning bagging, boosting, stacking, random forest	Ch 19.8	HW 9
11/16	Th	Linear Regression and Classification 1 loss, gradient descent, hill climbing	Ch 19.6
– November Break –
11/28	Tu	Linear Regression and Classification 2 perceptron, back-propagation, matrix multiplication	Ch 21.1
11/30	Th	Non-linear Models 1 hidden layer, activation function, softmax, convolution	Ch 21.2
12/05	Tu	Non-linear Models 2 k-nearest neighbors, kernel, support-vector machine	Ch 19.7	HW 10
12/07	Th	Ethics and Safety of AI topics TBD	Ch 27	Project 4 (Neural Nets)
12/12	Tu	No Class		HW 11
12/19	Tu	No Class		570 Project