Adaptive Agents with Reinforcement Learning and Internal Memory
Pier Luca Lanzi · 2000
Perceptual aliasing is a serious problem for adaptive agents. Internal memory is a promising approach to extend reinforcement learning algorithms to problems involving perceptual aliasing. In this paper we investigate the effectiveness of internal memory for tackling perceptual aliasing problems with adaptive agents and reinforcement learning. Specifically, we try to give a unified view of some interesting results that have been presented in different frameworks, i.e.: tabular reinforcement learning and learning classifier systems.