CS 371 - Introduction to Artificial Intelligence
Course Syllabus

Note: This syllabus is tentative and subject to change.  Each reading assignment should be completed before the class on the date indicated.  If a reading assigned in class does not match the reading assignment here, the reading assigned in class supersedes.  Unless otherwise noted, readings are from Russell & Norvig, "Artificial Intelligence: a modern approach, 4th ed.".

 
Class Date Topic Readings (parenthesized
reading are optional)
1 Introduction, "What is AI?"; Uninformed search: problem definition Course Information page; (Skim R&N Ch. 1, 2); R&N 3.1-3.3; Deep Blue article
2 Uninformed Search: breadth-first search, depth-first search, recursive depth-first search, depth-limited search, iterative-deepening depth-first search 3.4, HW1 Starter Code and Documentation, (Peg Solitaire Demonstration)
3 Uninformed Search: repeated state detection, in-class implementation   
4 Informed Search: heuristic search, best-first search, uniform cost search, greedy search rest of Ch. 3 except 3.5.5 (memory-bounded search)
5 Stochastic Local Search: hill descent with uphill step probability, simulated annealing, parameter tuning, demonstrations, challenge problem implementation 4.1, (4.2), Skim Science Simulated Annealing article
6 Stochastic Local Search: second challenge problem
7 Game-Tree Search: minimax, alpha-beta pruning 5.1-5.3; game-tree search slides; (FYS 187-4 videos: game trees, alpha-beta pruning)
8 Game-Tree Search: heuristic evaluation, in-class heuristic development 5.4; (FYS 187-4 videos: heuristic/static evaluation, heuristic features)
9 Game-Tree Search: time management H. Baier, M. Winands. Time Management for Monte-Carlo Tree Search in Go. (skim time management strategies)
10 Game-Tree Search: expectiminimax 5.5; Poker Squares rules/play grid, Learning and Using Hand Abstraction Values for Parameterized Poker Squares, (slides); (FYS 187-4 video: chance node, expectimax, and optimal Pig play reasoning)
11 Game-Tree Search: Monte Carlo Tree Search (MCTS), UCT Section 2 through 2.3 iof L. Kocsis, C. Szepesvari. Bandit based Monte-Carlo Planning.; Sections 1 (through 1.1) and 3.1 of C. Browne et al. A Survey of Monte Carlo Tree Search Methods
12 Machine Learning: Dynamic Programming "Solving the Dice Game Pig: an introduction to dynamic programming and value iteration", Sections 1-2; (For reference in class presentation: Pig slides)
13 Machine Learning: Dynamic Programming (cont.), Value Iteration "Solving the Dice Game Pig: an introduction to dynamic programming and value iteration", remainder; Russell & Norvig sections 17.1-17.2
14 Machine Learning: Value Iteration (cont.)  
15 In-class Midterm  
16 Q-Learning: definitions, backup diagram, update rule, application to Pig Solitaire section 6.5, of Reinforcement Learning: an introduction
17 Q-Learning: application to Pig Solitaire (continued)
18 Q-Learning: application to Pig Solitaire (completed)
19 Data Science and Machine Learning: Supervised, Unsupervised, and Reinforcment Learning ml-cs-core-1.ipynb
20 Data Science and Machine Learning: Underfitting/overfitting, bias-variance tradeoff, performance measures, train/validation/test datasets ml-cs-core-2.ipynb
21 Data Science and Machine Learning: Data Preprocessing, ML Concepts/Tradeoffs, Error Sources; Q-Learning Demonstrations ml-cs-core-3.ipynb
22 Data Science and Machine Learning: Data Preprocessing, ML Concepts/Tradeoffs, Error Sources; Q-Learning Demonstrations ml-cs-core-3.ipynb
23 Data Science and Machine Learning: Neural Networks, Q-Learning Demonstrations ml-cs-core-4.ipynb
24 Data Science and Machine Learning: Neural Networks (cont.), ML and Ethics ml-cs-core-4.ipynb
25 Probabilistic Reasoning: Bayesian Networks, Gibbs Sampling, a Markov Chain Monte Carlo (MCMC) algorithm for reasoning on Bayesian networks 14.1-14.2, An Introduction to Monte Carlo Techniques in AI - Part II, 14.5.2, Pearl, Judea. Probabilistic Reasoning in Intelligent Systems: Networks of Plausible Inference, Revised 2nd Printing, Morgan Kaufmann, San Francisco, California, USA, 1988, Table 1 (p. 197), section 4.4.3 (pp. 210-216, up to "Justifying the Computations")
26 Probabilistic Reasoning (cont.)  
27 Course Evaluations, ML Exercises  
28 4th Hour Project Demos  
    Final Exam Schedules on Registrar page under "Resources" (Fall, Spring)
Final: Monday, December 9th, 1:30 PM - 4:30 PM
Todd Neller
tneller@gettysburg.edu