CS 662 resources:

Solutions to midterm 1 can be found here . Total points: 150 Mean: 120 Median: 118 StDev: 15

Announcements:

Go here to sign up
for the cs662 mailing list. (you * must * be on this list - please use an
email address you check regularly.)

Regrading: If you feel that your homework was not graded correctly, you
should return it to me, along with a ** written ** explanation of the
error. I will evaluate this explanation and award any necessary points.

Midterms:

Solutions to midterm 2 can be found here . Total points: 150 Mean: 119 Median: 126

Solutions to midterm 1 can be found here . Total points: 150 Mean: 120 Median: 118 StDev: 15

Midterm 1 will be held Monday, October 13. It is in class, closed-book and closed-notes. You can find a list of potential topics here.

Readings:

A list of readings from AIMA, 2nd edition, by week.

- Week of 9/3: 1.1 (pp 1-5), 1.4 (pp 27-28) (the rest of chapter 1 is also very interesting reading)
- Week of 9/8: Chapter 2: 2.1-2.4, pp 32-54
- Week of 9/15: 3.1-3,2 (pp 59-68), 3.3-3.5 (pp 69-83)
- Week of 9/22: 4.1-4,3: (pp 94-118).
- Week of 9/27: provided below.
- Week of 10/6: 6.1-6.3 (pp 161-171)
- Week of 10/13: Midterm 1.
- Week of 10/20: Logic. 7.1-7.5 (197-220), Intro to Jess (10/20), plus intro to First-order logic (10/22) Required reading: R & N: Ch 8 (pp 240-268)
- Week of 10/27: More first-order logic, advanced Jess programming Required reading: R & N, Ch 9.
- Week of 11/3: Planning. Reading: R & N: 11.1-11.3, pp 375-394, decision trees. 18.1-18.3 pp 649-664.
- Week of 11/10: Probability. 13.1-13.6. pp 462-482. Bayesian learning. reading provided.
- Week of 11/17- Bayesian Networks. R & N 14.1-14.4. Utility theory and Decision Networks. 16.1-16.6.
- Week of 12/1: Neural Networks. R & N 20.5
- Week of 12/8: MDPs and Reinforcement Learning. R & N 17.1-17.3, 21.1-21.3

Lectures:

- Slides from Class 1: 9/3/03
- Slides from Class 2 & 3: Agents and Environments 9/8/03 and 9/10/03.
- Slides from Class 4 & 5: Uninformed Search 9/15/03 and 9/17/03
- Slides from Class 6: Heuristic Search 9/22/03
- Slides from Class 7: Local Search 9/22/03
- Slides from Class 8: Genetic Algorithms 9/29/03
- Slides from Class 9: Classifier Systems 10/1/03
- Slides from Class 10: Adversarial Search 10/6/03
- Slides from Class 11: Intro to Propositional Logic 10/15/03
- Slides from Class 12: Resolution in Propositional Logic and Introduction to Jess 10/20/03
- Slides from Class 13:
Introduction to First-order Logic 10/22/03
- Here is the Jess code and notes from the lecture.

- Slides from Class 14: Inference in First-order Logic 10/27/03
- Slides from Class 15: Planning 11/3/03
- Slides from Class 16: Decision Trees 11/5/03
- Slides from Class 17: Probability 11/10/03
- Slides from Class 18: Bayesian Learning 11/12/03
- Slides from Class 19: Bayesian Networks 11/17/03
- Slides from Class 20: Utility and Decision Networks 11/19/03
- Slides from Class 21: Neural Networks 12/1/03
- Slides from Class 22: Neural Networks II 12/3/03
- Slides from Class 23: Markov Decision Processes 12/8/03
- Slides from Class 24: Q-learning 12/10/03

Jess resources:

- Some notes on working with Jess.
- The Jess homepage
- The Jess manual. Note that we'll just be using the Jess language itself, not the Java extensions, so ignore all the stuff about Java.
- Jess functions These are the built-in functions that Jess has.
- The Jess reference manual A description of the language features. Again, skip over the Java stuff.
- The CLIPS User's guide. Jess is a Java port of CLIPS. This document is a user's guide to clips, with lots of examples. Almost all the language-relevant material in this manual applies to Jess.

- A Genetic Algorithm Tutorial Darrell Whitley. Statistics and Computing (4):65-85, 1994
- More will be provided in class.

Python links:

- The Python Home Page A wealth of information.
- A Python tutorial, written by Guido van Rossum, the original developer of Python.
- Another Python Intro
- Safari: O'Reilly's online library. Online access to all of O'Reilly's books, including their Python books. (Note: this is a subscription service, but well worth the $10/month, IMHO)
- A Python Quick Reference
- Python Online. This one's about Monty Python, rather than Python the language, but it's fun, and might explain why you see so many "spam" examples in Python tutorials.