Back to Search Start Over

Optimizing agent behavior over long time scales by transporting value

Authors :
Chia-Chun Hung
Timothy Lillicrap
Josh Abramson
Yan Wu
Mehdi Mirza
Federico Carnevale
Arun Ahuja
Greg Wayne
Source :
Nature Communications, Vol 10, Iss 1, Pp 1-12 (2019)
Publication Year :
2019
Publisher :
Nature Portfolio, 2019.

Abstract

People are able to mentally time travel to distant memories and reflect on the consequences of those past events. Here, the authors show how a mechanism that connects learning from delayed rewards with memory retrieval can enable AI agents to discover links between past events to help decide better courses of action in the future.

Subjects

Subjects :
Science

Details

Language :
English
ISSN :
20411723
Volume :
10
Issue :
1
Database :
Directory of Open Access Journals
Journal :
Nature Communications
Publication Type :
Academic Journal
Accession number :
edsdoj.2e805e7216b343f9a8b03a4f5ce02aad
Document Type :
article
Full Text :
https://doi.org/10.1038/s41467-019-13073-w