Reinforcement Learning With Human Advice: A Survey

Frontiers in Robotics and AI

Mohamed Chetouani¹  Anis Najar² 

[1] Institute for Intelligent Systems and Robotics, Sorbonne Université, CNRS UMR 7222, Paris, France;Laboratoire de Neurosciences Cognitives Computationnelles, INSERM U960, Paris, France;

关键词: advice-taking systems;    reinforcement learning;    interactive machine learning;    human-robot interaction;    unlabeled teaching signals;   

DOI  :  10.3389/frobt.2021.584075

来源: DOAJ

【 摘 要 】

In this paper, we provide an overview of the existing methods for integrating human advice into a reinforcement learning process. We first propose a taxonomy of the different forms of advice that can be provided to a learning agent. We then describe the methods that can be used for interpreting advice when its meaning is not determined beforehand. Finally, we review different approaches for integrating advice into the learning process.

【 授权许可】

Unknown