Bounded Finite State Controllers

Pascal Poupart
Department of Computer Science
University of Toronto
Toronto, ON M5S 3H5
email: ppoupart@cs.toronto.edu

Craig Boutilier
Department of Computer Science
University of Toronto
Toronto, ON M5S 3H5
email: cebly@cs.toronto.edu

Abstract
We describe a new approximation algorithm for solving partially observable MDPs. Our bounded policy iteration approach searches through the space of bounded-size, stochastic finite state controllers, combining several advantages of gradient ascent (efficiency, search through restricted controller space) and policy iteration (less vulnerability to local optima).

To appear, NIPS-03

Return to List of Papers