PSU CS410/510 CS
PSU/OCATE CS510 OC6
Advanced Artificial Intelligence: Combinatorial Search
Winter 2001
Syllabus

Note that this is a tentative syllabus, subject to change at any time.

When: 1800-2130 (6:00-9:00PM) Thursdays
Where: Urban Center room 205 (remote classroom)
Who: Bart Massey <bart@cs.pdx.edu>
Office Hours:by e-mail appointment
CRN: 40656(410), 40671(510), 45542(OCATE)

Summary

This course explores methods for the solution of constraint satisfaction and related problems using search techniques, in the context of real-world problems such as resource-bounded scheduling, enterprise planning, classical planning, and one- and two-player games. The emphasis is on coding projects, and on reading and reporting on selected literature.

The course is, of necessity, an overview of the area: there is far more material here than can be thoroughly covered in the time available. A student project and presentation will provide an opportunity for focus on a particular topic of individual interest.

Coursework

Each student will be expected to do all assigned reading, to present a paper in class, and to do several coding homework assignments culminating in a small final coding project.

In addition, graduate students will be expected to select, read, and present in class (all with my help) a paper of their choosing on a search-related topic.

This is a seminar-style course: I will attempt to substitute discussion for lecture to the extent that I find it feasible. It is important that students come to class prepared to discuss the material.

Prerequisites

Objectives

Upon the successful completion of this course, a student will:

Course Texts

The required course text will be

[Tsang]
Dr. Edward Tsang
Principles Of Constraint Satisfaction
Out-of-print MS, 1993
and is available at Clean Copy. If you have already purchased this text from this source between 2000-12-1 and 2000-1-10, it probably has every other page missing. If so, please return it to Clean Copy for a replacement.

An additional optional text,

[Garey-Johnson]
Michael Garey and David Johnson
Computers and Intractability: A Guide to the Theory of NP-Completeness
W H Freeman & Co.
ISBN: 0716710455
$29.95
will be very useful both in this course and throughout your career as a computer scientist or software engineer. I will try to provide photocopies of all required readings from this text, however.

Other papers will either be handed out in class or available on-line.

Other texts used in the course include:

[Ginsberg-book]
Matt Ginsberg
Essentials of Artificial Intelligence
Morgan Kaufman 1993
ISBN 1-55860-221-6
 
[Korf-book]
Dr. Richard Korf
Heuristic Search
Unpublished MS, 1999

Online Papers

[Gent-Walsh]
Ian P. Gent and Toby Walsh
The Search For Satisfaction
Web Document, 2000
http://dream.dai.ed.ac.uk/group/tw/sat/sat-survey2.ps
[PS] [PDF]
 
[Gupta-Nau]
Naresh Gupta and Dana S. Nau
On The Complexity Of Blocks-World Planning
Artificial Intelligence 56(2-3):223-254, August 1992
[PS] [PDF]
[Korf-CKK]
Richard E. Korf
A Complete Anytime Algorithm for Number Partitioning
University of California CS TR 970020
[PS] [PDF]
(Also published as Artificial Intelligence (105)1-2, pp. 133-135, 1998)
[Minton-Johnston]
Steven Minton, Mark D. Johnston, Andrew B. Philips, and Philip Laird
Minimizing Conflicts: A Heuristic Repair Method for Constraint-Satisfaction and Scheduling Problems [PS] [PDF]

Links

Mailing List

There is one special e-mail address associated with the course. <cs510cs-discuss@cs.pdx.edu> is the course mailing list (using majordomo). Subscribe to this by sending an e-mail message to majordomo@cs.pdx.edu with subscribe cs510cs-discuss in the body, and use it for class discussions and the like. I will also announce important changes to the web page to this list.

Tentative Schedule

Readings are for the week after the lecture, (but of course feel free to do them earlier).

[PS] [PDF]
Date Topics Readings HW
1/11 Introduction To Search [Ginsberg-book] ch. 1, 2 (handout)  
1/18 Constraint Satisfaction Problems [Tsang] ch. 1, 2 HW 1a (due 1/31)
1/25 NP-Completeness [PS] [PDF] [Garey-Johnson] ch. 1, 2 (handout)  
2/1 Good Complete Search Methods [PS] [PDF] [Tsang] ch. 2.3, 5.1-5.2, 10.1-10.2.3, [Korf-CKK] HW 1b (due 2/21)
2/8 Incomplete Search Methods and SAT [PS] [PDF] [Tsang] ch. 8, [Gent-Walsh] HW 1c (due 2/28)
2/15 Pathfinding Search, Adversary Search [PS] [PDF] [Korf-book] ch. 3 (except 3.7.4-), ch. 7  
2/22 More About CSPs, Search Software Engineering [PS] [PDF] [Tsang] ch. 3-6 (where relevant)  
3/1 Planning and Scheduling [PS] [PDF] [Gupta-Nau], [Minton-Johnston] Final Project (due 3/21)
Final Presentation (in-class 3/15)
3/8 More About Planning And SchedulingTBA  
3/15 Student Presentations    

Previous Material

Old homework assignments and some notes on them are online. There are also some links to online papers that were part of the course reading.


Last Modified: 2001/01/15
Bart Massey, <bart@cs.pdx.edu>