ABSTRACT
Intelligent systems must often reason with partial or corrupted information, due to noisy sensors, limited representation capabilities, and inherent problem complexity. Gathering new information and reasoning with existing information comes at a computational or physical cost. This paper presents a formalism to model systems that solve logical reasoning problems in the presence of uncertainty and priced information. The system is modeled a decision-making agent that moves in a probabilistic belief space, where each information-gathering or computation step changes the belief state. This forms a Markov decision process (MDP), and the belief-optimal system operates according to the belief-space policy that optimizes the MDP. This formalism makes the strong assertion that belief-optimal systems solve the reasoning problem at minimal expected cost, given the background knowledge, sensing capabilities, and computational resources available to the system. Furthermore, this paper argues that belief-optimal systems are more likely to avoid overfitting to benchmarks than benchmark-optimized systems. These concepts are illustrated on a variety of toy problems as well as a path optimization problem encountered in motion planning.
- D. Berry and B. Fristedt. Bandit Problems. Chapman and Hall, 1985.Google ScholarCross Ref
- C. Boutilier, T. Dean, and S. Hanks. Decision-theoretic planning: Structural assumptions and computational leverage. Journal of Artificial Intelligence Research, 11:1--94, 1999.Google ScholarDigital Library
- B. Burns and O. Brock. Single-query motion planning with utility-guided random trees. In IEEE Int. Conf. Rob. Aut., 2007.Google ScholarCross Ref
- V. A. Cicirello and S. F. Smith. The max k-armed bandit: A new model of exploration applied to search heuristic selection. In AAAI, 2005. Google ScholarDigital Library
- D. Fouskakis and D. Draper. Comparing stochastic optimization methods for variable selection in binary outcome prediction, with application to health policy. Journal of the American Statistical Association, 03(484):1367--1381, 2008.Google ScholarCross Ref
- R. Geraerts and M. Overmars. Clearance based path optimization for motion planning. In IEEE Int. Conf. Rob. Aut., New Orleans, LA, 2004.Google ScholarCross Ref
- R. Geraerts and M. H. Overmars. Creating high-quality paths for motion planning. Intl. J. of Rob. Res., 26(8):845--863, 2007. Google ScholarDigital Library
- R. Greiner, R. Hayward, M. Jankowska, and M. Molloy. Finding optimal satisficing strategies for and-or trees. Artificial Intelligence, 170:19--58, 2006. Google ScholarDigital Library
- K. Hauser. Motion Planning for Legged and Humanoid Robots. PhD thesis, Stanford University, 2008. Google ScholarDigital Library
- D. Hsu, J. Latombe, and H. Kurniawati. On the probabilistic foundations of probabilistic roadmap planning. Int. J. Rob. Res., 25(7):627--643, 2006. Google ScholarDigital Library
- D. Hsu, G. Sánchez-Ante, and Z. Sun. Hybrid prm sampling with a cost-sensitive adaptive strategy. In IEEE Int. Conf. Rob. Aut., pages 3885--3891, Barcelona, Spain, 2005.Google ScholarCross Ref
- M. Kearns, Y. Mansour, and A. Y. Ng. Approximate planning in large pomdps via reusable trajectories. In Advances in Neural Information Processing Systems 12. MIT Press, 2000.Google Scholar
- H. Kurniawati, D. Hsu, and W. Lee. Sarsop: Efficient point-based pomdp planning by approximating optimally reachable belief spaces. In Proc. Robotics: Science and Systems, 2008.Google ScholarCross Ref
- S. M. LaValle and J. J. Kuffner, Jr. Rapidly-exploring random trees: progress and prospects. In WAFR, 2000.Google Scholar
- O. Madani, S. Hanks, and A. Condon. On the undecidability of probabilistic planning and infinite-horizon partially observable markov decision problems. In AAAI/IAAI, pages 541--548, 1999. Google ScholarDigital Library
- C. H. Papadimitriou and J. N. Tsitsiklis. The complexity of markov chain decision processes. Mathematics of Operations Research, 12(3):441--450, 1987. Google ScholarDigital Library
- A. Rubinstein. Modeling Bounded Rationality. MIT Press, 1998.Google ScholarCross Ref
- G. Sánchez and J.-C. Latombe. On delaying collision checking in PRM planning: Application to multi-robot coordination. Int. J. of Rob. Res., 21(1):5--26, 2002.Google ScholarCross Ref
- J. F. Traub, H. W. G. W., and Wasilkowski. Information-Based Complexity. Academic Press, New York, 1988. Google ScholarDigital Library
- G. Wilfong. Motion planning in the presence of movable obstacles. In Fourth Annual Symposium on Computational Geometry, pages 279--288, Urbana-Champaign, IL, 1988. Google ScholarDigital Library
- M. Zlochin, M. Birattari, N. Meuleau, and M. Dorigo. Model-based search for combinatorial optimization: A critical survey. Annals of Operations Research, 131(1--4):373--395, October 2004.Google ScholarCross Ref
Recommendations
Linguistic hesitant fuzzy multi-criteria decision-making method based on evidential reasoning
Linguistic hesitant fuzzy sets LHFSs, which can be used to represent decision-makers’ qualitative preferences as well as reflect their hesitancy and inconsistency, have attracted a great deal of attention due to their flexibility and efficiency. This ...
Default reasoning through belief revision strategy
AAAI'87: Proceedings of the sixth National conference on Artificial intelligence - Volume 1The thesis of this paper is that default reasoning can be accomplished rather naturally if an appropriate strategy of belief revision is employed. The idea is based on the premise that new beliefs introduced into a situation change the structure of ...
Comments