Gradient Descent for General Reinforcement Learning

November 1998

Gradient Descent for General Reinforcement Learning

Authors:

Leemon Baird and Andrew Moore
Conference Paper
Proceedings of (NeurIPS) Neural Information Processing Systems

Abstract:

A simple learning rule is derived, the VAPS algorithm, which can be instantiated to generate a wide range of new reinforcement-learning algorithms. These algorithms solve a number of open problems, define several new approaches to reinforcement learning, and unify different approaches to reinforcement learning under a single theory. These algorithms all have guaranteed convergence, and include modifications of several existing algorithms that were known to fail to converge on simple MDPs. These include Q-learning, SARSA, and advantage learning. In addition to these value-based algorithms it also generates pure policy-search reinforcement-learning algorithms, which learn optimal policies without learning a value function. In addition, it allows policy-search and value-based algorithms to be combined, thus unifying two very different approaches to reinforcement learning into a single Value and Policy Search (VAPS) algorithm. And these algorithms converge for POMDPs without requiring a proper belief state. Simulations results are given, and several areas for future research are discussed.
@conference{Baird-1998-16688,
author = {Leemon Baird And Andrew Moore},
title = {Gradient Descent for General Reinforcement Learning},
booktitle = {Proceedings of (NeurIPS) Neural Information Processing Systems},
year = {1998},
month = {November},
pages = {968 - 974},
}
Copyright notice: This material is presented to ensure timely dissemination of scholarly and technical work. Copyright and all rights therein are retained by authors or by other copyright holders. All persons copying this information are expected to adhere to the terms and constraints invoked by each author's copyright. These works may not be reposted without the explicit permission of the copyright holder.