Potential-based multiobjective reinforcement learning approaches to low-impact agents for AI safety

Vamplew, P, Foale, C, Dazeley, Richard and Bignold, A 2021, Potential-based multiobjective reinforcement learning approaches to low-impact agents for AI safety, Engineering Applications of Artificial Intelligence, vol. 100, pp. 1-16, doi: 10.1016/j.engappai.2021.104186.

Attached Files
Name Description MIMEType Size Downloads

Title Potential-based multiobjective reinforcement learning approaches to low-impact agents for AI safety
Author(s) Vamplew, P
Foale, C
Dazeley, RichardORCID iD for Dazeley, Richard orcid.org/0000-0002-6199-9685
Bignold, A
Journal name Engineering Applications of Artificial Intelligence
Volume number 100
Article ID 104186
Start page 1
End page 16
Total pages 16
Publisher Elsevier
Place of publication Amsterdam, The Netherlands
Publication date 2021-04
ISSN 0952-1976
Keyword(s) Safe reinforcement learning
Multiobjective reinforcement learning
AI safety
Potential-based rewards
Low-impact agents
Reward engineering
Side-effects
Language eng
DOI 10.1016/j.engappai.2021.104186
Indigenous content off
Field of Research 08 Information and Computing Sciences
09 Engineering
HERDC Research category C1 Refereed article in a scholarly journal
Persistent URL http://hdl.handle.net/10536/DRO/DU:30147988

Connect to link resolver
 
Unless expressly stated otherwise, the copyright for items in DRO is owned by the author, with all rights reserved.

Versions
Version Filter Type
Citation counts: TR Web of Science Citation Count  Cited 0 times in TR Web of Science
Scopus Citation Count Cited 0 times in Scopus
Google Scholar Search Google Scholar
Access Statistics: 8 Abstract Views, 1 File Downloads  -  Detailed Statistics
Created: Thu, 11 Feb 2021, 09:43:04 EST

Every reasonable effort has been made to ensure that permission has been obtained for items included in DRO. If you believe that your rights have been infringed by this repository, please contact drosupport@deakin.edu.au.