In-context Exploration-Exploitation for Reinforcement Learning

arXiv:2403.06826v1 Announce Type: new
Abstract: In-context learning is a promising approach for online policy learning of offline reinforcement learning (RL) methods, which can be achieved at inference time without gradient optimization. However, this method is hindered by significant computational costs resulting from the gathering of large training trajectory sets and the need to train large Transformer models. We address this challenge by introducing an In-context Exploration-Exploitation (ICEE) algorithm, designed to optimize the efficiency of in-context policy learning. Unlike existing models, …

abstract arxiv computational context costs cs.ai cs.lg exploitation exploration gradient however in-context learning inference offline optimization policy reinforcement reinforcement learning stat.ml train training trajectory type