Symmetric Primal-Dual Approximate Linear Programming for Factored MDPs

Dmitri A. Dolgov, and Edmund H. Durfee

In Proceedings of the Ninth International Symposiums on Artificial Intelligence and Mathematics (ISAIM 2006). January 2006.

Abstract
A weakness of classical Markov decision processes is that they scale very poorly due to the flat state-space representation. Factored MDPs address this representational problem by exploiting problem structure to specify the transition and reward functions of an MDP in a compact manner. However, in general, solutions to factored MDPs do not retain the structure and compactness of the problem representation, forcing approximate solutions, with approximate linear programming (ALP) emerging as a very promising MDP-approximation technique. To date, most of ALP work has focused on the primal-LP formulation, while the dual LP, which forms the basis for solving constrained Markov problems, has received much less attention. We show that a straightforward linear approximation of the dual optimization variables is problematic, because some of the required computations cannot be carried out efficiently. Nonetheless, we develop a composite approach that symmetrically approximates the primal and dual optimization variables (effectively approximating both the objective function and the feasible region of the LP) that is computationally feasible and suitable for solving constrained MDPs. We empirically show that this new ALP formulation also performs well on unconstrained problems.


BibTex
@inproceedings{ dolgov06alp,
   paperID   = "ISAIM-06",
   month     = "January",
   author    = "Dmitri A. Dolgov and Edmund H. Durfee",
   booktitle = "Proceedings of the Ninth International Symposiums on Artificial Intelligence and Mathematics (ISAIM 2006)",
   address   = "Florida",
   title     = "Symmetric Primal-Dual Approximate Linear Programming for Factored {MDP}s",
   year      = "2006"
}


Download:
pdf [pdf]        ps [ps.gz]