A Linearly Relaxed Approximate Linear Program for Markov Decision Processes
Abstract
Approximate linear programming (ALP) and its variants have been widely applied to Markov Decision Processes (MDPs) with a large number of states. A serious limitation of ALP is that it has an intractable number of constraints, as a result of which constraint approximations are of interest. In this paper, we define a linearly relaxed approximation linear program (LRALP) that has a tractable number of constraints, obtained as positive linear combinations of the original constraints of the ALP. The main contribution is a novel performance bound for LRALP.
- Publication:
-
arXiv e-prints
- Pub Date:
- April 2017
- DOI:
- 10.48550/arXiv.1704.02544
- arXiv:
- arXiv:1704.02544
- Bibcode:
- 2017arXiv170402544L
- Keywords:
-
- Computer Science - Systems and Control
- E-Print:
- 23 pages, 2 figures, submitted to IEEE TAC