-
Tytuł:
-
Learning offline: memory replay in biological and artificial reinforcement learning.
-
Autorzy:
-
Roscow EL; Centre de Recerca Matemàtica, Bellaterra, Spain. Electronic address: .
Chua R; McGill University and Mila, Montréal, Canada.
Costa RP; Bristol Computational Neuroscience Unit, Intelligent Systems Lab, Department of Computer Science, University of Bristol, Bristol, UK.
Jones MW; School of Physiology, Pharmacology and Neuroscience, University of Bristol, Bristol, UK.
Lepora N; Department of Engineering Mathematics and Bristol Robotics Laboratory, University of Bristol, Bristol, UK.
-
Źródło:
-
Trends in neurosciences [Trends Neurosci] 2021 Oct; Vol. 44 (10), pp. 808-821. Date of Electronic Publication: 2021 Sep 01.
-
Typ publikacji:
-
Journal Article; Research Support, Non-U.S. Gov't; Review
-
Język:
-
English
-
Imprint Name(s):
-
Publication: Barking : Elsevier Applied Science Publishing
Original Publication: Amsterdam, New York, Elsevier/North-Holland Biomedical Press.
-
MeSH Terms:
-
Artificial Intelligence*
Hippocampus*
Humans ; Machine Learning ; Reinforcement, Psychology ; Reward
-
Grant Information:
-
202810/Z/16/Z United Kingdom WT_ Wellcome Trust; 109070/Z/15/A United Kingdom WT_ Wellcome Trust
-
Contributed Indexing:
-
Keywords: Q-learning; computation; deep neural networks; hippocampus; reward
-
Entry Date(s):
-
Date Created: 20210905 Date Completed: 20211020 Latest Revision: 20211020
-
Update Code:
-
20240105
-
DOI:
-
10.1016/j.tins.2021.07.007
-
PMID:
-
34481635
-
Learning to act in an environment to maximise rewards is among the brain's key functions. This process has often been conceptualised within the framework of reinforcement learning, which has also gained prominence in machine learning and artificial intelligence (AI) as a way to optimise decision making. A common aspect of both biological and machine reinforcement learning is the reactivation of previously experienced episodes, referred to as replay. Replay is important for memory consolidation in biological neural networks and is key to stabilising learning in deep neural networks. Here, we review recent developments concerning the functional roles of replay in the fields of neuroscience and AI. Complementary progress suggests how replay might support learning processes, including generalisation and continual learning, affording opportunities to transfer knowledge across the two fields to advance the understanding of biological and artificial learning and memory.
Competing Interests: Declaration of interests The authors declare no competing interests in relation to this work.
(Copyright © 2021 Elsevier Ltd. All rights reserved.)