Efficient Approximate Linear Programming for Factored MDPs

Feng Chen,Qiang Cheng,Jianwu Dong,Zhaofei Yu,Guojun Wang,Wenli Xu
DOI: https://doi.org/10.1016/j.ijar.2015.06.002
IF: 4.452
2015-01-01
International Journal of Approximate Reasoning
Abstract:Factored Markov Decision Processes (MDPs) provide a compact representation for modeling sequential decision making problems with many variables. Approximate linear programming (LP) is a prominent method for solving factored MDPs. However, it cannot be applied to models with large treewidth due to the exponential number of constraints. This paper proposes a novel and efficient approximate method to represent the exponentially many constraints. We construct an augmented junction graph from the factored MDP, and represent the constraints using a set of cluster constraints and separator constraints, where the cluster constraints play the role of reducing the number of constraints, and the separator constraints enforce the consistency of neighboring clusters so as to improve the accuracy. In the case where the junction graph is tree-structured, our method provides an equivalent representation to the original constraints. In other cases, our method provides a good trade-off between computation and accuracy. Experimental results on different models show that our algorithm performs better than other approximate linear programming algorithms on computational cost or expected reward.
What problem does this paper attempt to address?