Reinforcement Learning methods for controlling stochastic processes typically assume a small and discrete action space. While continuous action spaces are quite common in real-wor...
Abstract. We investigate the problem of using function approximation in reinforcement learning where the agent’s policy is represented as a classifier mapping states to actions....
This paper investigates a novel model-free reinforcement learning architecture, the Natural Actor-Critic. The actor updates are based on stochastic policy gradients employing Amari...
Production scheduling, the problem of sequentially con guring a factory to meet forecasted demands, is a critical problem throughout the manufacturing industry. The requirement of...
Jeff G. Schneider, Justin A. Boyan, Andrew W. Moor...
Abstract. In this paper we study distributed online learning of locomotion gaits for modular robots. The learning is based on a stochastic approximation method, SPSA, which optimiz...