Efficient learning in linearly solvable MDP models

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Scopus citations

Abstract

Linearly solvable Markov Decision Process (MDP) models are a powerful subclass of problems with a simple structure that allow the policy to be written directly in terms of the uncontrolled (passive) dynamics of the environment and the goals of the agent. However, there have been no learning algorithms for this class of models. In this research, we develop a robust learning approach to linearly solvable MDPs. To exploit the simple solution for general problems, we show how to construct passive dynamics from any transition matrix, use Bayesian updating to estimate the model parameters and apply approximate and efficient Bayesian exploration to speed learning. In addition, we reduce the computational cost of learning using intermittent Bayesian updating and policy solving. We also gave a polynomial theoretical time complexity bound for the convergence of our learning algorithm, and demonstrate a linear bound for the subclass of the reinforcement learning problems with the property that the transition error depends only on the agent itself. Test results for our algorithm in a grid world are presented, comparing our algorithm with the BEB algorithm. The results showed that our algorithm learned more than the BEB algorithm without losing convergence speed, so that the advantage of our algorithm increased as the environment got more complex. We also showed that our algorithm's performance is more stable after convergence. Finally, we show how to apply our approach to the Cellular Telephones problem by defining the passive dynamics.

Original languageEnglish (US)
Title of host publicationIJCAI 2013 - Proceedings of the 23rd International Joint Conference on Artificial Intelligence
Pages248-253
Number of pages6
StatePublished - Dec 1 2013
Event23rd International Joint Conference on Artificial Intelligence, IJCAI 2013 - Beijing, China
Duration: Aug 3 2013Aug 9 2013

Publication series

NameIJCAI International Joint Conference on Artificial Intelligence
ISSN (Print)1045-0823

Other

Other23rd International Joint Conference on Artificial Intelligence, IJCAI 2013
CountryChina
CityBeijing
Period8/3/138/9/13

Fingerprint Dive into the research topics of 'Efficient learning in linearly solvable MDP models'. Together they form a unique fingerprint.

Cite this