Automated Reinforcement Learning (AutoRL)

A Survey and Open Problems

verfasst von

Jack Parker-Holder, Raghu Rajan, Xingyou Song, André Biedenkapp, Yingjie Miao, Theresa Eimer, Baohe Zhang, Vu Nguyen, Roberto Calandra, Aleksandra Faust, Frank Hutter, Marius Lindauer

Abstract

The combination of Reinforcement Learning (RL) with deep learning has led to a series of impressive feats, with many believing (deep) RL provides a path towards generally capable agents. However, the success of RL agents is often highly sensitive to design choices in the training process, which may require tedious and error-prone manual tuning. This makes it challenging to use RL for new problems and also limits its full potential. In many other areas of machine learning, AutoML has shown that it is possible to automate such design choices, and AutoML has also yielded promising initial results when applied to RL. However, Automated Reinforcement Learning (AutoRL) involves not only standard applications of AutoML but also includes additional challenges unique to RL, that naturally produce a different set of methods. As such, AutoRL has been emerging as an important area of research in RL, providing promise in a variety of applications from RNA design to playing games, such as Go. Given the diversity of methods and environments considered in RL, much of the research has been conducted in distinct subfields, ranging from meta-learning to evolution. In this survey, we seek to unify the field of AutoRL, provide a common taxonomy, discuss each area in detail and pose open problems of interest to researchers going forward.

Organisationseinheit(en)

Fachgebiet Maschinelles Lernen

Externe Organisation(en)

Albert-Ludwigs-Universität Freiburg
University of Oxford
Google Research
Amazon Australia
Meta AI
Bosch Center for Artificial Intelligence (BCAI)

Typ

Artikel

Journal

Journal of Artificial Intelligence Research

Band

74

Seiten

517-568

Anzahl der Seiten

52

ISSN

1076-9757

Publikationsdatum

01.06.2022

Publikationsstatus

Veröffentlicht

Peer-reviewed

Ja

Elektronische Version(en)

https://doi.org/10.48550/arXiv.2201.03916 (Zugang: Offen)
https://doi.org/10.1613/jair.1.13596 (Zugang: Geschlossen)