Abstract
Any anytime algorithm used for decision-making should have the property that it considers the most important aspects of the decision problem first. In this way, the algorithm can first eliminate disastrous decisions and recognize particularly advantageous decisions, considering finer details if time permits. We view planning as a decision-making process and discuss the design of anytime algorithms for decision-theoretic planning. In particular, we present an anytime decision-theoretic planning algorithm that uses abstraction to focus attention first on those aspects of a planning problem that have the highest impact on expected utility. We discuss control schemes for refining this behavior and methods for automatically creating good abstractions.
- {Boutilier and Dearden, 1994} C. Boutilier and R. Dearden. Using abstractions for decision-theoretic planning with time constraints. In Proceedings of the Twelfth National Conference on Artificial Intelligence, pages 1016--1022, Seattle, July 1994. Google ScholarDigital Library
- {Dean and Boddy, 1988} T. Dean and M. Boddy. An analysis of time-dependent planning. In Proceedings of the Seventh National Conference on Artificial Intelligence, pages 49--54, Saint Paul, MN, August 1988.Google Scholar
- {Dean et al., 1993} T. Dean, L. Pack Kaelbling, J. Kirman, and A. Nicholson. Planning with deadlines in stochastic domains. In Proceedings of the Eleventh National Conference on Artificial Intelligence, pages 574--579, July 1993.Google Scholar
- {Dearden and Boutilier, 1994} R. Dearden and C. Boutilier. Integrating planning and execution in stochastic domains. In Proceedings of the Tenth Conference on Uncertainty in Artificial Intelligence, pages 162--169, Seattle, July 1994.Google ScholarDigital Library
- {Doan and Haddawy, 1995} A. Doan and P. Haddawy. Generating macro operators for decision-theoretic planning. In Working Notes of the AAAI Spring Symposium on Extending Theories of Action, pages 68--73, Stanford, March 1995.Google Scholar
- {Doan, 1995} A. Doan. An abstraction-based approach to decision-theoretic planning for partially observable metric domains. Master's thesis, Dept. of EE & CS, University of Wisconsin-Milwaukee, December 1995.Google Scholar
- {Drummond and Bresina, 1990} M. Drummond and J. Bresina. Anytime synthetic projection: Maximizing the probability of goal satisfaction. In Proceedings of the Eighth National Conference on Artificial Intelligence, pages 138--144, Boston, MA, July 1990.Google Scholar
- {Finigan, 1995} M. G. Finigan. Knowledge acquisition for decision-theoretic planning. Master's thesis, Dept. of EE & CS, Univ. of Wisconsin-Milwaukee, August 1995.Google Scholar
- {Haddawy and Suwandi, 1994} P. Haddawy and M. Suwandi. Decision-theoretic refinement planning using inheritance abstraction. In Proceedings, Second International Conference on AI Planning Systems, pages 266--271, June 1994.Google ScholarDigital Library
- {Haddawy et al., 1995} P. Haddawy, A. Doan, and R. Goodwin. Efficient decision-theoretic planning: Techniques and empirical analysis. In Proceedings of the Eleventh Conference on Uncertainty in Artificial Intelligence, pages 229--236, Montreal, August 1995. Google ScholarDigital Library
- {Hanks et al., 1993} S. Hanks, M. E. Pollack, and P. R. Cohen. Benchmarks, test beds, controlled experimentation, and the design of agent architectures. AI Magizine, 14(4):17--42, Winter 1993. Google ScholarDigital Library
- {Hanks, 1990} S. Hanks. Projecting Plans for Uncertain Worlds. PhD thesis, Yale University, January 1990. Google ScholarDigital Library
- {Horvitz, 1988} E. J. Horvitz. Reasoning under varying and uncertain resource constraints. In Proceedings of the Seventh National Conference on Artificial Intelligence, pages 111--116, Saint Paul, MN, August 1988.Google ScholarDigital Library
- {Keeney and Raiffa, 1976} R. L. Keeney and H. Raiffa. Decisions with Multiple Objectives: Preferences and Value Tradeoffs. Wiley, New York, 1976.Google Scholar
- {Thiebaux et al., 1994} S. Thiebaux, J. Hertzberg, W. Shoaff, and M. Schneider. A stochastic model of actions and plans for anytime planning under uncertainty. International Journal of Intelligent Systems, 1994.Google Scholar
Index Terms
- Focusing attention in anytime decision-theoretic planning
Recommendations
Decision-theoretic planning: structural assumptions and computational leverage
Planning under uncertainty is a central problem in the study of automated sequential decision making, and has been addressed by researchers in many different fields, including AI planning, decision analysis, operations research, control theory and ...
Comments