George Mason University
School of Information Technology and Engineering
DEPARTMENT OF COMPUTER SCIENCE
INFT910 - Special Topics in Artificial Intelligence: Intelligent Agents - Fall '96
Thursday 7:20pm-10:00pm, Robinson B205
News |
Description|
Readings
This page last updated on 8/19/96.
Professor Gheorghe
Tecuci
703-993-1722
tecuci@ gmu.edu
(for fastest service prefix the subject of your email with INFT910)
Course office hour Thursday 6:10pm-7:10pm; e-mail all times
ST2 Room 421
NEWS:
To get a computer account at SITE: (1) Obtain an OSF1 account from UCIS.
(2) Login to OSF1 and type "sitereg". You will be prompted for
your GMU id and will be allowed to create a SITE account on line.
Other relevant class news will be posted here.
Back to the top.
DESCRIPTION :
INFT910 - Special Topics in Artificial Intelligence: Intelligent Agents
Prerequisite: a graduate level course in artificial intelligence
or permission of instructor.
An agent is a system situated within and part of an environment that senses that environment and acts on it, over time, in pursuit of its own agenda and so as to effect what it senses in the future (Franklin & Graesser, 96). Intelligent software agents are AI programs that can assist users in various ways. For instance, they can perform tasks on the user's behalf, they can advise, train or teach the user, they can monitor events for the user, or help different users collaborate. This course will review and analyze the basic approaches to building intelligent software agents, as well as representative examples of current agent applications. Particular attention will be given to learning agents that can be directly trained by a user to perform specific tasks, and that can learn from their experience. The course will also discuss open issues, current trends and frontier research in the area of intelligent agents.
The course will consist of a sequence of lectures given by the instructor and student presentations. Students will select papers from a provided list and will present critical reviewes and analyses of representative intelligent agents. Students may also experiment with or enhance the agent building software developed in the Learning Agents Laboratory.
Back to the top.
READINGS:
Recommended readings for each topic and week will appear in this space.
Most of the papers will be available on the Internet.
Overview of intelligent agents
- Stan Franklin and Art Graesser,
Is it an Agent, or Just a Program?, Proc. of the Third International Workshop on Agent Theories, Architectures, and Languages, Springer-Verlag, 1996.
- Stuart Russell and Peter Norvig, Artificial Intelligence: A Modern Approach, Chapter 2: Intelligent Agents, Englewood Cliffs, NJ: Prentice Hall, 1995.
- IBM, Intelligent Agent Strategy
- Pattie Maes, Intelligent Software, Scientific American, Vol. 273, No.3, pp. 84-86, September 1995.
-
Intelligent Agents = Stupid Humans?, A web-based debate between Pattie Maes and Jaron Lanier which deals with the feasibility and desirability of intelligent agents (sponsored by the Wired magazine).
- Riecken, Doug, Intelligent agents, Communications of the ACM (special issue on Intelligent Agents), July 1994, Vol.37, No.7, pp. 18-21.
- Minsky, Marvin and Riecken, Doug, A conversation with Marvin Minsky about agents, Communications of the ACM (special issue on Intelligent Agents), July 1994, Vol.37, No.7, pp. 22-29.
- Slides, ( Postscript
)
Foundations for building intelligent agents
- G. Tecuci, Machine Learning and Knowledge Acquisition: Integrated Approaches, ( Compressed Postscript ), Tutorial Notes, The International Joint Conference on Artificial Intelligence, 143 pages, IJCAI-95, Montreal, Canada, August 1995.
The Disciple approach to building intelligent agents
- Tecuci G. and Kodratoff Y., Apprenticeship Learning in Imperfect Domain Theories, ( Compressed Postscript ), in Kodratoff Y. and Michalski R.S. (eds), Machine Learning: An Artificial Intelligence Approach, vol III, pp. 514-551, Morgan Kaufmann, 1990.
- Tecuci G., Hieb M.H and Dybala T., Scaling Up the Disciple Approach, ( Compressed Postscript ),
Figure 2, ( Compressed Postscript ),
Figure 3, ( Compressed Postscript ),
Draft, 1996.
- Tecuci G. and Hieb M.H, Teaching Intelligent Agents: The Disciple Approach, ( Compressed Postscript ), International Journal of Human-Computer Interaction, vol.8, no.3, pp.259-285, 1996.
- Dybala T, Tecuci G and Keeling H., Building Disciple Agents: Selected Validation and Verification Issues, ( Compressed Postscript ), AAAI-96 Workshop on Validation and Verification of Knowledge-Based Systems, Portland, Oregon, August 1996.
Softbots for the Internet
- O. Etzioni, D. Weld, A Softbot-Based Interface to the Internet, ( Postscript
), Communications of the ACM, Vol. 37, No. 7, July 1994.
- O. Etzioni, Intelligence without Robots (A Reply to Brooks), ( Compressed Postscript ),AI Magazine, Dec. 93.
- O. Etzioni, Moving Up the Information Food Chain: Deploying Softbots on the World Wide Web ( Compressed Postscript ), Invited talk, Proc. AAAI-96, pp.1322-1326.
Agents that reduce work and information overload
- P. Maes, Agents that Reduce Work and Information Overload, ( HTML
), Communications of the ACM, Vol. 37, No. 7, July 1994.
- Collaborative Interface Agents
(Postscript or
HTML)
Proceedings of AAAI '94 Conference, Seattle, Washington, August 1994.
Yezdi Lashkari,
Max Metral, and
Pattie Maes
- P. Maes and R.Kozierok, Learning Interface Agents, Proceedings of AAAI'93, 8 pages.
Learning personal assistants
- T.Mitchell, R.Caruana, D.Freitag, J.McDermott, D.Zabowski, Experience with a Learning Personal Assistant, ( Compressed Postscript
), Communications of the ACM, Vol. 37, No. 7, July 1994.
- T.Mitchell, D.Freitag, T.Joachims, R.Armstrong, D.Zabowski, The WebWatcher Project, ( HTML
)
Agent-based software engineering
- M.R.Genesereth, Software Agents, Communications of the ACM, Vol. 37, No. 7, pp.48-53, p.147, July 1994.
Programming agents by example
- H. Lieberman, Programming by example home page, ( HTML
)
- Cypher, Allen and Smith, David C. "KidSim: End User Programming of Simulations".
In Proceedings of CHI, 1995 (Denver, May 7 - 11). ACM, New York,
1995, pp. 27-34. [Paper Session at CHI]
- Smith, David C., and Cypher, Allen. "KidSim: Child Constructible Simulations". In Proceedings of the Imagina '95 Conference, Monte-Carlo, February 1995, pp. 87-99.
- Smith, D.C., Cypher,
A. & Spohrer, J. "KidSim: Programming Agents Without a Programming Language". In Communications of the ACM, 37(7), July 1994, pp. 54 - 67.
- A User Interface for Knowledge Acquisition from Video
(HTML,
RTF,
Postscript,
Quicktime movie)
Proceedings of AAAI '94 Conference, Seattle, Washington, August 1994.
Henry Lieberman
- Mondrian, a Teachable Graphical Editor
(HTML,
RTF,
Postscript,
Quicktime movie)
Watch What I Do,
Allen Cypher, ed.,
MIT Press, 1993.
Henry Lieberman
Teaching agents that learn
- T. Selker, Coach: a teaching agent that Learns, Communications of the ACM, Vol. 37, No. 7, pp.92-99, July 1994.
Architecture of integrated agents
- D. Riecken, M: an architecture of integrated agents, Communications of the ACM, Vol. 37, No. 7, pp.107-116, p.146, July 1994.
Enabling agents to work together
- R.V.Guha, D.B.Lenat, Cyc: Enabling agents to work together, Communications of the ACM, Vol. 37, No. 7, pp.127-142, July 1994.
Disciple Toolkit Related Projects
- Develop the Rule Editor
- Test and integrate the Exception Handler into the Disciple toolkit
- Extend the Rule Refinement tool
- Extend and improve the Explanation Grapher
- Develop Agent's capabilities of explaining its actions
- Develop a Demo in the History Domain and its documentation in a hypertext format
- Use the Disciple toolkit to build a learning agent in a domain of your choice
Back to the top.