Generalizing Multi-Step Inverse Models for Representation Learning to Finite-Memory POMDPs

arXiv:2404.14552v1 Announce Type: new
Abstract: Discovering an informative, or agent-centric, state representation that encodes only the relevant information while discarding the irrelevant is a key challenge towards scaling reinforcement learning algorithms and efficiently applying them to downstream tasks. Prior works studied this problem in high-dimensional Markovian environments, when the current observation may be a complex object but is sufficient to decode the informative state. In this work, we consider the problem of discovering the agent-centric state in the more challenging …

abstract agent algorithms arxiv challenge cs.ai cs.lg current environments information key memory prior reinforcement reinforcement learning representation representation learning scaling state tasks them type