T/Th 2:00pm-3:15pm, Smith Hall 102a
|Instructor: Keith Decker|
|Office: Room 206, The Green House (77 E. Delaware)|
|Hours: T/Th 1:00pm-2:00pm|
|Email: email@example.com (preferred contact method)|
This course introduces students to the fields of Multi-Agent Systems and Distributed Artificial Intelligence, which deal with the issues that arise when groups or societies of autonomous agents (usually computer programs but sometimes people too) interact to solve interrelated problems. These agents may be self-interested or cooperating to solve a shared problem. Important issues include reasoning about the knowledge & beliefs of other agents; high-level communication & negotiation protocols; the organization, coordination, and control of complex, distributed computation. Applications include internet information gathering, electronic commerce, and workflow management in areas such as finance, bioinformatics, etc...
Multiagent Systems: A Modern Approach to Distributed Artificial Intelligence,
edited by Gerhard Weiss, MIT Press, 1999. ISBN 0-262-23203-0
Seminar with practical implementation project. Most of the class will be discussion of the text we are reading, and later perhaps some selected external papers. While early-on I and guest lecturers will lead/teach class, later on we will all tradeoff the role of discussion leader.
We will study both the theory and implementation of multi-agent systems, using our local DECAF (Distributed, Environment-Centered Agent Framework) multi-agent toolkit for implementation. We will do a small "get acquainted" programming assignment, and then some group projects (including bio-informatics, for the people who are interested).
This sylabus is preliminary, and subject to change to meet your needs as students. In particular, the end sections of the course will need to be changed, depending on what people find interesting.
|8/28||Intelligent Agents||Chapter 1|
DECAF Programming: An Introduction[McGeary]
|9/11||Multiagent Systems||Chapter 2|
|9/18||Agent Communication Languages||FIPA:
Communicative Act Library XC00037G.pdf
Interaction Protocol Library XC00025D.pdf
you can browse some others HERE.
|9/25||Distributed Problem Solving||Chapter 3|
|10/2||Coordination||Coordination Technologies for
DAI [Nick Jennings] (PDF VERSION)
|10/9||Coordination||Coordinating Mutually Exclusive
Resources Using GPGP [PDF]
|10/16||Coordination||Analyzing the need for meta-level communication
|10/18|| GPGP Evolution
|10/23||Decision Making||Chapter 5|
|11/6||Distributed Constraint Satisfaction||Chapter 4.2|
|11/8||Agent SWE|| N. R. Jennings (2000) "On Agent-Based Software Engineering"
Artificial Intelligence, 117 (2) 277-296.
|11/13||Agent SWE||M. J. Wooldridge and N. R. Jennings (1998)
"Pitfalls of Agent-Oriented Development"
(Agents-98), Minneapolis, USA, 385-391.
M. Wooldridge, N. R. Jennings, and D. Kinny (2000)
|11/15||Application: Flex. Manufac.||S. Bussmann, K. Schild: Self-Organizing Manufacturing Control:
An Industrial Application of Agent Technology.
In Proc. of the Fourth Int. Conf. on Multi-Agent
Systems, pages 87 -- 94, Boston, MA, USA, 2000.
[Comparision of OO approach] O. Lassila and S. Smith,
Agent SWE & Coordination
|E. M. Atkins, T. F. Abdelzaher, K. G. Shin, and E. H. Durfee, Planning and Resource Allocation for Hard Real-time, Fault-Tolerant Plan Execution, Journal of Autonomous Agents and Multi-Agent Systems, V4#1/2, 2001.
Munindar P. Singh. Synthesizing Coordination Requirements for Heterogeneous Autonomous Agents (pdf) Autonomous Agents and Multi-Agent Systems. volume 3, number 2, June 2000, pages 107-132.
|11/29||DCS||Makoto Yokoo & K. Hirayama. Algorithms for Distributed Constraint Satisfaction: A Review, Autonomous Agents and Multi-Agent Systems, Vol.3, No.2, pp.189--212, 2000.
Bradley J. Clement, Edmund H. Durfee. Performance of Coordinating Concurrent Hierarchical Planning Agents Using Summary Information. Proceedings of the Seventh International Workshop on Agent Theories, Architectures, and Languages, 202-216, 2000.
|12/4||Dist. Planning & Scheduling||Paper 1|
Grades will be awarded for one homework, final project, class participation, and discussion leadership.
Since this is a graduate seminar/discussion class, it is vital that material be read before class. In general it will NOT be the subject of any lecture at all! Students may prepare answers to some of the discussion questions at the end of the chapters (when indicated beforehand).