Proceedings

Prioritized Goal Decomposition of Markov Decision Processes: Toward a Synthesis of Classical and Decision Theoretic Planning

Download BibTeX Cite

Proceedings of the Fifteenth International Joint Conference on Artificial Intelligence. Planning and Scheduling. Pages 1156-1161.

© 1997 International Joint Conferences on Artificial Intelligence

Abstract

We describe an approach to goal decomposition for a certain class of Markov decision processes (MOPs) . An abstraction mechanism is used to generate abstract MOPs associated with different objectives, and several methods for merging the policies for these different objectives are considered. In one technique, causal (least-commitment) structures are generated for abstract policies and plan merg ing techniques, exploiting the relaxation of policy commitments reflected in this structure, are used to piece the results into a single policy. Abstract value functions provide guidance if plan repair is needed. This work makes some first steps toward the synthesis of classical and decision theoretic planning methods .

  1. R. E. Bellman. Dynamic Programming . Princeton, 1957.
  2. C. Boutilier, R. Dearden, and M. Goldszmidt. Exploiting structure in policy construction. In /JCA/-95, pp.ll 04-1111, Montreal, 1995. IJCAI
  3. M. E. Bratman. Intentions, Plans, and Practical Reason. Har- vard, 1987.
  4. T. Dean, L. P. Kaelbling, J. Kirman, and A. Nicholson. Plan- ning with deadlines in stochastic domains. In AAAI-93, pp.574-579, Washington, D.C., 1993.
  5. T. Dean and K. Kanazawa. A model for reasoning about per- sistence and causation. Comp. Intel., 5(3): 142-150, 1989.
  6. R. Dearden and C. Boutilier. Abstraction and approximate de- cision theoretic planning. Artif. Intel., 89:219-283, 1997.
  7. R. A. Howard. Dynamic Programming and Markov Pro- cesses. MIT Press. Cambridge, 1960.
  8. R. L. Keeney and H. Raiffa. Decisions with Multiple Objec- tives: Preferences and Value Trade-offs. Wiley, New York, 1978.
  9. N. Kushmerick, S. Hanks, and D. Weld. for probabilistic least-commitment planning. pp.l 073-1078, Seattle, 1994. An algorithm In AAA/-94,
  10. D. McAIIester and D. Rosenblitt. Systematic nonlinear plan- ning. In-AAA/-91 , pp.634-639,Anaheim, 1991.
  11. J. S. Penberthy and D. S. Weld. UCPOP: A sound, complete, partial order planner for ad!. In KR-92, 1992.
  12. M. L. Puterman. Markov Decision Processes: Discrete Stochastic Dynamic Programming . Wiley, NY, 1994.•
  13. D. S. Weld. An introduction to least commitment planning. AI Magazine, Winter 1994:27-61, 1994.
  14. Q. Yang, D. S. Nau, and J. Hendler. Merging seperately gen- erated plans in limited domains. Comp. Intel., 8(4), 1992.

References were extracted automatically from the PDF and may contain errors.

Welcome Future Conferences AI Journal Division IJCAI Proceedings About IJCAI IJCAI Awards