CS 463G Artificial Intelligence Syllabus

Class page

Time and Place: 1--1:50 PM, 337 CB

Professor: Dr. J. Goldsmith
Office: 763E Anderson Hall, Phone: 257-4245
Office Hours: Mondays 10 AM, Wednesdays 11 AM, and by appointment. Email questions strongly encouraged and answered.

Course Description:

The topics covered in this course will be:

• intelligent agents
  
• search and game-playing
  
• logical systems
  
• planning systems
  
• uncertainty---probability theory, inference, and decision theory
  

Prereqs: CS 315 and CS 375, and engineering or graduate standing. You should know how to program, be familiar with basic algorithms and data structures, especially for graphs and trees, and be familiar with propositional and predicate logic.

Textbook: Artificial Intelligence: A Modern Approach, 2nd Edition by Stuart Russell and Peter Norvig. Prentice Hall, 2003.

Grading:

There will be assignments every other week, due Fridays at the beginning of class. Problems will be posted by the previous Wednesday on the web and program assignments two weeks prior to the due date. The lowest homework grade will be dropped. Illegible work will not be graded. Plagiarized work will be penalized for all parties, according to University regulations. There will be one midterm and one final. These will not be exams, unless y'all really want exams.

Assignments (problems and programs) will be 65% of your grade, the midterm project will be 15% and the final 20%. The midterm project will be completed by October 19th; a project proposal is due on September 21st. The final project is due by Dec. 14th at 1pm, although any in-class presentations must be scheduled during a class period. The final project proposal is due November 9.

Those taking CS 463G for graduate credit (in any department except CS) will have one additional assignment of a paper or presentation.

READ THIS:

Attendance in class and section is very strongly encouraged.

Copying of homework from other students or from other sources is strictly prohibited. Obtaining a solution from another source without citing the source is plagiarism. You are encouraged to visit Dr. Goldsmith or your T.A. in their office hours or to send them email if you are stuck on homework problems. You do not need an appointment for regularly scheduled hours.

Outcomes and assessments

Students will learn basic concepts in logic and artificial intelligence. In particular, the students will be able to:

• implement and analyze search algorithms: depth-first search, breadth-first search, best-first search, A*, min-max and and alpha-beta pruning
• use logic to represent knowledge and reason from it
• formalize planning problems and apply algorithms to compute plans
• formalize domain knowledge as Bayesian networks
• design and reason about agents

Week by Week Course Outline:
Consider this syllabus a first approximation.

Date	Topic	Chapter	Assignment
	INTRODUCTION
Aug. 26-8	Intro to course, AI, agents	1
	SEARCH
Aug. 31-Sept. 4	Agents, uninformed search	2,3	Puzzle, part 1
Sept. 9-11	Informed search	3,4
Sept. 14-18	More searching, constraint satisfaction	4,5	IDA* program
Sept. 21	MIDTERM PROJECT PROPOSAL DUE
Sept. 21-25	More constraint algorithms		SAT programs
No class on Monday, Sept. 28: Jewish High Holy Day
Sept. 30-Oct. 2	Games	6
	LOGICAL SYSTEMS
Oct. 5-9	Propositional and predicate logics (CS 375)	7, 8	Games programs
Oct. 12-16	Wumpus world, situation calculus	7,8
Oct. 19	MIDTERM PROJECT DUE
Oct. 19--23	Inference, resolution, Prolog	9
Oct. 26--30	Knowledge representation	10	Prolog program
	PLANNING AND UNCERTAINTY
Nov. 2-6	Planning	11
Nov. 9	FINAL PROJECT PROPOSAL DUE
Nov. 9-13	Uncertainty: Probability theory, Na\"ive Bayes classifiers	13
Nov. 16-20	Bayesian networks	14--17
Nov. 23	Planning under uncertainty
Nov. 25-27	THANKSGIVING BREAK (Class does not meet)
Nov. 30-Dec. 4	Machine learning	18	MCMC program
Dec. 7-11	Topics in AI	20
Dec. 14	FINAL PROJECT DUE: 1 PM