Publikation
Symbolic Dynamic Programming for Continuous State and Observation POMDPs
Zahra Zamani; Scott Sanner; Pascal Poupart; Kristian Kersting
In: Peter L. Bartlett; Fernando C. N. Pereira; Christopher J. C. Burges; Léon Bottou; Kilian Q. Weinberger (Hrsg.). Advances in Neural Information Processing Systems 25: 26th Annual Conference on Neural Information Processing Systems 2012. Neural Information Processing Systems (NeurIPS-2012), December 3-6, Lake Tahoe, Nevada, USA, Pages 1403-1411, Curran Associates, Inc. 2012.
Zusammenfassung
Point-based value iteration (PBVI) methods have proven extremely effective for finding (approximately) optimal dynamic programming solutions to partiallyobservable Markov decision processes (POMDPs) when a set of initial belief states is known. However, no PBVI work has provided exact point-based backups for both continuous state and observation spaces, which we tackle in this paper. Our key insight is that while there may be an infinite number of observations, there are only a finite number of continuous observation partitionings that are relevant for optimal decision-making when a finite, fixed set of reachable belief states is considered. To this end, we make two important contributions:(1) we show how previous exact symbolic dynamic programming solutions for continuous state MDPs can be generalized to continuous state POMDPs with discrete observations, and (2) we show how recently developed symbolic integration methods allow this solution to be extended to PBVI for continuous state and observation POMDPs with potentially correlated, multivariate continuous observation spaces.