Modules

All material on this website is © Devika Subramanian, 2007-2017. Please request permission from devika@rice.edu for use of the material, and please acknowledge this site in your material.

This course is modular in design. All chapter readings and questions below are from the course textbook “Artificial Intelligence: A Modern Approach”, by Russell and Norvig, 3rd edition. All quizzes and assignments are due on Canvas at 8 pm on the dates specified below.

Syllabus

Week 1: Introduction and search (8/21-8/25)

• Lectures:
  • Introduction What can AI do for you?
    • Read Chapter 1.1-1.5, 2.1-2.3
  • Search 1 Problem solving as path finding
    • Read Chapter 2.4-2.5, 3.1-3.4
    • basic search algorithms
• Term project (final report due 1 December, 8 pm on Canvas)
• Assignment 1 (due 4 September, 8 pm on Canvas; moved to 8 September)
• Quiz 1 (due 25 August at 8 pm on Canvas)

Week 2: Heuristic search and local search (8/28-9/1)

• Lectures:
  • Search 2 Algorithms for path finding in graphs: UCS, A*
    • Read Chapter 3.5-3.7
  • Search 3 Local search: HC, GSAT and GAs
    • Read Chapter 4.1-4.2, 4.5
• Assignment 1 (due 4 September, 8 pm on Canvas; moved to 8 September)
• Quiz 2 (due 1 September at 8 pm on Canvas; moved to 8 September)

Week 3: Adversarial search and game playing (9/4-9/8)

• Lectures:
  • Games 1 Adversarial search, minimax and alpha-beta pruning
    • Read Chapter 5.1-5.3
  • Games 2 Stochastic adversaries, expectimax,evaluation function learning
    • Read Chapter 5.4-5.5, 5.7-5.9
• Assignment 1 (due 4 September, 8 pm on Canvas; moved to 8 September)
• Assignment 2 (due 21 September, 8 pm on Canvas)
• Term project Pacwar blog I (due 15 September, 8 pm at blogs.rice.edu)
• Quiz 3 (due 15 September, 8 pm on Canvas)

Week 4: Game theory and MDPs (9/11-9/15)

• Lectures:
  • Games 3 Game theory, Nash equilibria, mechanism design
    • Read Chapter 17.5-17.6
  • MDP 1 Markov decision processes
    • Read Chapter 17.1-17.2
• Assignment 2 (due 21 September, 8 pm on Canvas)
• Assignment 3 (due 2 October, 8 pm on Canvas)
• Quiz 4 (due 22 September, 8 pm on Canvas)

Week 5: Markov decision processes (9/18-9/22)

• Lectures:
  • MDP 2 Policy evaluation, policy improvement
    • Read Chapter 17.2-17.3
  • MDP 3 Policy iteration, value iteration
    • Read Chapter 17.2-17.3
• Assignment 3 (due 2 October, 8 pm on Canvas)
• Assignment 4 (due 18 October, 8 pm on Canvas)
• Quiz 5 (due 29 September, 8 pm on Canvas)

Week 6: Constraint satisfaction (9/25-9/29)

• Lectures:
  • CSP 1 Modeling problems as CSPs, backtracking search
    • Read Chapter 6.1-6.3
  • CSP 2 Dynamic variable and value ordering
    • Read Chapter 6.3-6.4
• Assignment 3 (due 2 October, 8 pm on Canvas)
• Assignment 4 (due 18 October, 8 pm on Canvas)
• Quiz 6 (due 29 September, 8 pm on Canvas)

Week 7: Bayesian networks (10/2-10/6)

• Lectures:
  • BN 0 Introduction to probability
    • Read Chapter 13.1-13.5
  • BN 1 Bayesian networks, Conditional independence
    • Read Chapter 14.1-14.4
• Assignment 4 (due 18 October, 8 pm on Canvas)
• Midterm on 5 October 7pm-10pm, location TBA

Week 8: BNs and probabilistic reasoning (10/9-10/13)

• Lectures:
  • BN 2 Approximate inference
    • Read Chapter 14.5
  • BN 2 Gibbs sampling
    • Read Chapter 14.5
• Assignment 4 (due 18 October, 8 pm on Canvas)
• Assignment 5 (due 27 October, 8 pm on Canvas)
• Quiz 7 (due 13 October, 8 pm on Canvas)

Week 9: Temporal models and HMMs (10/16-10/20)

• Lectures:
  • BN 3 Sequential models, exact inference in HMMs
    • Read Chapter 15.1-15.3
  • BN 3 Approximate inference in HMMs: Particle filtering
    • Read Chapter 15.4-15.6
• Assignment 5 (due 27 October, 8 pm on Canvas)
• Term project Pacwar blog II (due 16 October, 8 pm at blogs.rice.edu)
• Quiz 8 (due 20 October, 8 pm on Canvas)

Week 10: Decision networks and supervised learning (10/23-10/27)

• Lectures:
  • DN 1 Decision networks
    • Read Chapter 16.1-16.6
  • SL 1 Supervised learning: linear classification, perceptrons, naive Bayes
    • Read Chapter 18.1, 18.4, 18.5,  18.6-18.7
• Assignment 5 (due 27 October, 8 pm on Canvas)
• Assignment 6 (due 10 November, 8 pm on Canvas)
• Quiz 9 (due 27 October, 8 pm on Canvas)

Week 11: Supervised learning (10/30-11/3)

• Lectures:
  • SL 2 Nearest neighbors, kernel methods
    • Read Chapter 18.6-18.8
  • SL 3 Decision trees, MLPs
    • Read Chapter 18.4, 18.7
• Assignment 6 (due 10 November, 8 pm on Canvas)
• Quiz 10 (due 3 November, 8 pm on Canvas)

Week 12: Supervised learning (11/6-11/10)

• Lectures:
  • SL 4 Deep learning
    • Read Chapter 18.7
  • SL 5 Ensemble models: bagging, boosting
    • Read Chapter 18.4, 18.7
• Assignment 6 (due 10 November, 8 pm on Canvas)
• Assignment 7 (due 24 November, 8 pm on Canvas)
• Quiz 11 (due 10 November, 8 pm on Canvas)

Week 13: Unsupervised learning (11/13-11/17)

• Lectures:
  • UL 1 Clustering algorithms
    • Read Chapter 20.1-20.2
  • UL 2 PCA, ICA and tSNE
    • Read Chapter 20.3-20.4
• Assignment 7 (due 24 November, 8 pm on Canvas)
• Term project Pacwar blog III (due 15 November, 8 pm at blogs.rice.edu)
• Quiz 12 (due 17 November, 8 pm on Canvas)

Week 14: Reinforcement learning (11/20-11/24)

• Lectures:
  • RL 1 Reinforcement learning: the basics
    • Read Chapter 21.1-21.3
  • No class, Thanksgiving break
• Term project Draft of final report (due 22 November, 8 pm on Canvas)

Week 15: Reinforcement learning and wrap-up (11/27-12/1)

• Lectures:
  • RL 2 Practical reinforcement learning
    • Read Chapter 21.5-21.7
  • Wrap up Review of course
    • Read Chapter 26, Chapter 27
• Term project Pacmite due (due 29 November, 12 noon on piazza as private post)
• Term project PacWar tournament (29 November, 7 pm at McMurtry Aud. DH)
• Term project PacWar final report (1 December, 8 pm on Canvas)