Abstract
Most work on adaptive agents have a simple, single layerarchitecture. However, most agent architectures support three levels ofknowledge and control: a reflex level for reactive responses, a deliberatelevel for goal-driven behavior, and a reflective layer for deliberateplanning and problem decomposition. In this paper we explore agentsimplemented in Soar that behave and learn at the deliberate and reflectivelevels. These levels enhance not only behavior, but also adaptation. Theagents use a combination of analytic and empirical learning, drawing from avariety of sources of knowledge to adapt to their environment. We hypothesize that complete, adaptive agents must be able to learn across all three levels.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
DeJong, G. A case study of explanation-based control. In Proceedings of the Twelfth International Workshop on Machine Learning, pages 167–175, Los Altos, CA, 1995. Morgan Kaufmann Publishers, Inc.
Dietterich, T.G. and N. S. Flann. Explanation based learning and reinforcement learning: A unified view. In Proceedings of the Twelfth International Workshop on Machine Learning, pages 176–184, Los Altos, CA, 1995. Morgan Kaufmann Publishers, Inc.
Doorenbos, Robert B. Matching 100,000 learned rules. In Proceedings of the Eleventh National Conference on Artificial Intelligence, pages 290–296. AAAI Press, 1993.
Forgy, C.L. Rete: A fast algorithm for the many pattern / many object pattern match problem. Artificial Intelligence, 19(1):17–38, 1982.
Gil, Yolanda. A domain-independent framework for effective experimentation in planning. In Proceedings of the International Machine Learning Workshop, pages 13–17, 1991.
Huffman, S.B. and J. E. Laird. Flexibly instructable agents. Journal of Artificial Intelligence Research, 3:271–324, 1995.
Laird, J.E. Recovery from incorrect knowledge in Soar. In Proceedings of the Seventh National Conference on Artificial Intelligence, pages 618–623, August 1988.
Laird, J.E., A. Newell, and P. S. Rosenbloom. Soar: Anarchitecture for general intelligence. Artificial Intelligence, 33(1):1–64, 1987.
Laird, J.E. and P. S. Rosenbloom. Integrating execution, planning, and learning in Soar for external environments. In Proceedings of the Eighth National Conference on Artificial Intelligence, pages 1022–1029, July 1990.
Lehman, J.F., J. E. Laird, and P. S. Rosenbloom. A gentle introduction to Soar, an architecture for human cognition. In Invitation to Cognitive Science, Volume 4. MIT Press, 1996.
Lehman, J.F., A. Newell, T. Polk, and R. Lewis. The role of language in cognition: A computational inquiry. In Conceptions of the Human Mind. Lawrence Erlbaum Associates, Inc, 1993.
Lenat, D.B. EURISKO: A program that learns new heuristics and domain concepts. The nature of heuristics III: Program design and results. Artificial Intelligence, 20:61–98, 1983.
Miller, C.M. A model of concept acquisition in the context of a unified theory of cognition. PhD thesis, The University of Michigan, Deptment of Electrical Engineering and Computer Science, 1993.
Miller, C.S. and J. E. Laird. Accounting for graded performance within a discrete search framework. Cognitive Science, 20:499–537, 1996.
Mitchell, T.M., R. M. Keller, and S. T. Kedar-Cabelli. Explanation-based generalization: A unifying view. Machine Learning, 1, 1986.
Mitchell, T.M. and S. B. Thrun. Explanation based learning: A comparison of symbolic and neural network approaches. In Proceedings of the Tenth International Workshop on Machine Learning, pages 197–204, Los Altos, CA, 1993. Morgan Kaufmann Publishers, Inc.
Newell, A. Unified Theories of Cognition. Harvard University Press, Cambridge, Massachusetts, 1990.
Ourston, D. and R. J. Mooney. Changing the rules: A comprehensive approach to theory refinement. In Proceedings of the National Conference on Artificial Intelligence, pages 815–820. AAAI/MIT Press, 1990.
Pearson, D.J. Learning Procedural Knowledge in Complex Environments. PhD thesis, Department of Electrical Engineering and Computer Science, University of Michigan, 1996.
Pearson, D.J., S. B. Huffman, M. B. Willis, J. E. Laird, and R. M. Jones. A symbolic solution to intelligent real-time control. Robotics and Autonomous Systems, 11:279–291, 1993.
Pearson, D.J. and J. E. Laird. Toward incremental knowledge correction for agents in complex environments. In S. Muggleton, D. Michie, and K. Furukawa, editors, Machine Intelligence, volume 15. Oxford University Press, 1996.
Rosenbloom, P.S., J. E. Laird, and A. Newell. The Soar Papers: Research on Integrated Intelligence. MIT Press, 1993.
Rosenbloom, P.S., J. E. Laird, A. Newell, and R. McCarl. A preliminary analysis of the Soar architecture as a basis for general intelligence. Artificial Intelligence, 47:75–111, 1991.
Tambe, M., W. L. Johnson, R. M. Jones, F. Koss, J. E. Laird, P. S. Rosenbloom, and K. Schwamb. Intelligent agents for interactive simulation environments. AI Magazine, 16(1):15–39, 1995.
Winograd, T. Understanding Natural Language. Academic Press, New York, 1972.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Laird, J.E., Pearson, D.J. & Huffman, S.B. Knowledge-directed Adaptation in Multi-level Agents. Journal of Intelligent Information Systems 9, 261–275 (1997). https://doi.org/10.1023/A:1008606203724
Issue Date:
DOI: https://doi.org/10.1023/A:1008606203724