Toward high-performance, memory-efficient, and fast reinforcement learning-Lessons from decision neuroscience

Jee Hang Lee; Ben Seymour; Joel Z Leibo; Su Jin An; Sang Wan Lee

doi:10.1126/scirobotics.aav2975

Toward high-performance, memory-efficient, and fast reinforcement learning-Lessons from decision neuroscience

Sci Robot. 2019 Jan 16;4(26):eaav2975. doi: 10.1126/scirobotics.aav2975.

Authors

Jee Hang Lee^{1

2}, Ben Seymour^{3

4

5}, Joel Z Leibo⁶, Su Jin An¹, Sang Wan Lee^{7

2

8}

Affiliations

¹ Department of Bio and Brain Engineering, KAIST, Daejeon, Republic of Korea.
² KAIST Institute for Health Science and Technology, Daejeon, Republic of Korea.
³ Computational and Biological Learning Laboratory, Department of Engineering, University of Cambridge, Trumpington Street, Cambridge CB2 1PZ, UK. bjs49@cam.ac.uk sangwan@kaist.ac.kr.
⁴ Brain Information Communication Research Laboratory Group, Advanced Telecommunications Research Institute International, Kyoto, Japan.
⁵ Center for Information and Neural Networks, National Institute of Information and Communications Technology, 1-4 Yamadaoka, Suita, Osaka 565-0871, Japan.
⁶ DeepMind, London, UK.
⁷ Department of Bio and Brain Engineering, KAIST, Daejeon, Republic of Korea. bjs49@cam.ac.uk sangwan@kaist.ac.kr.
⁸ KAIST Institute for Artificial Intelligence, Daejeon, Republic of Korea.

PMID: 33137757
DOI: 10.1126/scirobotics.aav2975

Abstract

Recent insights from decision neuroscience raise hope for the development of intelligent brain-inspired solutions to robot learning in real dynamic environments full of noise and unpredictability.

Publication types

Review
Research Support, Non-U.S. Gov't

Abstract

Publication types

Grants and funding