Fast value iteration for goal-directed Markov decision processes

Authors:
Nevin L. Zhang;Weihoag Zhang
Affiliations:
Department of Computer Science, Hong Kong University of Science & Technology;Department of Computer Science, Hong Kong University of Science & Technology
Venue:
UAI'97 Proceedings of the Thirteenth conference on Uncertainty in artificial intelligence
Year:
1997

Citing 8
Cited 0

Dynamic programming: deterministic and stochastic models

Dynamic programming: deterministic and stochastic models
Introduction to algorithms

Introduction to algorithms
A model for reasoning about persistence and causation

Computational Intelligence
Planning and control

Planning and control
Decomposition Techniques for Planning in Stochastic Domains

Decomposition Techniques for Planning in Stochastic Domains
Exploiting structure in policy construction

IJCAI'95 Proceedings of the 14th international joint conference on Artificial intelligence - Volume 2
Planning with deadlines in stochastic domains

AAAI'93 Proceedings of the eleventh national conference on Artificial intelligence
Model reduction techniques for computing approximately optimal solutions for Markov decision processes

UAI'97 Proceedings of the Thirteenth conference on Uncertainty in artificial intelligence

Quantified Score

Hi-index	0.00

Visualization

Abstract

Planning problems where effects of actions are non-deterministic can be modeled as Markov decision processes. Planning problems axe usually goal-directed. This paper proposes several techniques for exploiting the goal-directedness to accelerate value iteration, a standard algorithm for solving Markov decision processes. Empirical studies have shown that the techniques can bring about significant speedups.