Please use this identifier to cite or link to this item: http://hdl.handle.net/1942/37946
Full metadata record
DC FieldValueLanguage
dc.contributor.authorVamplew, Peter-
dc.contributor.authorSmith, Benjamin J.-
dc.contributor.authorKallstrom, Johan-
dc.contributor.authorRamos, Gabriel-
dc.contributor.authorRadulescu, Roxana-
dc.contributor.authorRoijers, Diederik M.-
dc.contributor.authorHayes, Conor F.-
dc.contributor.authorHeintz, Fredrik-
dc.contributor.authorMannion, Patrick-
dc.contributor.authorLIBIN, Pieter-
dc.contributor.authorDazeley, Richard-
dc.contributor.authorFoale, Cameron-
dc.date.accessioned2022-09-01T09:13:19Z-
dc.date.available2022-09-01T09:13:19Z-
dc.date.issued2022-
dc.date.submitted2022-08-16T11:29:58Z-
dc.identifier.citationAUTONOMOUS AGENTS AND MULTI-AGENT SYSTEMS, 36 (2) (Art N° 41)-
dc.identifier.urihttp://hdl.handle.net/1942/37946-
dc.description.abstractThe recent paper "Reward is Enough" by Silver, Singh, Precup and Sutton posits that the concept of reward maximisation is sufficient to underpin all intelligence, both natural and artificial, and provides a suitable basis for the creation of artificial general intelligence. We contest the underlying assumption of Silver et al. that such reward can be scalar-valued. In this paper we explain why scalar rewards are insufficient to account for some aspects of both biological and computational intelligence, and argue in favour of explicitly multi-objective models of reward maximisation. Furthermore, we contend that even if scalar reward functions can trigger intelligent behaviour in specific cases, this type of reward is insufficient for the development of human-aligned artificial general intelligence due to unacceptable risks of unsafe or unethical behaviour.-
dc.description.sponsorshipOpen Access funding enabled and organized by CAUL and its Member Institutions. This research was supported by funding from the Flemish Government under the “Onderzoeksprogramma Artifciële Intelligentie (AI) Vlaanderen” program, and by the National Cancer Institute of the U.S. National Institutes of Health under Award Number 1R01CA240452-01A1. The content is solely the responsibility of the authors and does not necessarily represent the ofcial views of the National Institutes of Health or of other funders. Pieter J.K. Libin acknowledges support from the Research Foundation Flanders (FWO, fwo.be) (postdoctoral fellowship 1242021N). Johan Källström and Fredrik Heintz were partially supported by the Swedish Governmental Agency for Innovation Systems (Grant NFFP7/2017- 04885), and the Wallenberg Artifcial Intelligence, Autonomous Systems and Software Program (WASP) funded by the Knut and Alice Wallenberg Foundation. Conor F. Hayes is funded by the National University of Ireland Galway Hardiman Scholarship. Gabriel Ramos was partially supported by FAPERGS (Grant 19/2551-0001277-2) and FAPESP (Grant 2020/05165-1).-
dc.language.isoen-
dc.publisherSPRINGER-
dc.rightsOpen Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.-
dc.subject.otherScalar rewards-
dc.subject.otherVector rewards-
dc.subject.otherArtificial general intelligence-
dc.subject.otherReinforcement learning-
dc.subject.otherMulti-objective decision making-
dc.subject.otherMulti-objective reinforcement learning-
dc.subject.otherSafe and ethical AI-
dc.titleScalar reward is not enough: a response to Silver, Singh, Precup and Sutton (2021)-
dc.typeJournal Contribution-
dc.identifier.issue2-
dc.identifier.volume36-
local.bibliographicCitation.jcatA1-
dc.description.notesVamplew, P (corresponding author), Federat Univ Australia, Ballarat, Vic, Australia.-
dc.description.notesp.vamplew@federation.edu.au; benjsmith@gmail.com;-
dc.description.notesjohan.kallstrom@liu.se; gdoramos@unisinos.br; roxana.radulescu@vub.be;-
dc.description.notesdiederik.roijers@vub.be; c.hayes13@nuigalway.ie; fredrik.heintz@liu.se;-
dc.description.notespatrickmannion@nuigalway.ie; pieter.libin@vub.be;-
dc.description.notesrichard.dazeley@deakin.edu.au; c.foale@federation.edu.au-
local.publisher.placeVAN GODEWIJCKSTRAAT 30, 3311 GZ DORDRECHT, NETHERLANDS-
local.type.refereedRefereed-
local.type.specifiedArticle-
local.bibliographicCitation.artnr41-
dc.identifier.doi10.1007/s10458-022-09575-5-
dc.identifier.isi000826149200001-
dc.contributor.orcidVamplew, Peter/0000-0002-8687-4424-
local.provider.typewosris-
local.description.affiliation[Vamplew, Peter; Foale, Cameron] Federat Univ Australia, Ballarat, Vic, Australia.-
local.description.affiliation[Smith, Benjamin J.] Univ Oregon, Ctr Translat Neurosci, Eugene, OR 97403 USA.-
local.description.affiliation[Kallstrom, Johan; Heintz, Fredrik] Linkoping Univ, Linkoping, Sweden.-
local.description.affiliation[Ramos, Gabriel] Univ Vale Rio dos Sinos, Sao Leopoldo, RS, Brazil.-
local.description.affiliation[Radulescu, Roxana] Vrije Univ Brussel, AI Lab, Brussels, Belgium.-
local.description.affiliation[Roijers, Diederik M.; Libin, Pieter J. K.] Vrije Univ Brussel, Brussels, Belgium.-
local.description.affiliation[Roijers, Diederik M.] HU Univ Appl Sci Utrecht, Utrecht, Netherlands.-
local.description.affiliation[Hayes, Conor F.; Mannion, Patrick] Natl Univ Ireland Galway, Galway, Ireland.-
local.description.affiliation[Libin, Pieter J. K.] Univ Hasselt, Hasselt, Belgium.-
local.description.affiliation[Libin, Pieter J. K.] Katholieke Univ Leuven, Leuven, Belgium.-
local.description.affiliation[Dazeley, Richard] Deakin Univ, Geelong, Vic, Australia.-
local.uhasselt.internationalyes-
item.fullcitationVamplew, Peter; Smith, Benjamin J.; Kallstrom, Johan; Ramos, Gabriel; Radulescu, Roxana; Roijers, Diederik M.; Hayes, Conor F.; Heintz, Fredrik; Mannion, Patrick; LIBIN, Pieter; Dazeley, Richard & Foale, Cameron (2022) Scalar reward is not enough: a response to Silver, Singh, Precup and Sutton (2021). In: AUTONOMOUS AGENTS AND MULTI-AGENT SYSTEMS, 36 (2) (Art N° 41).-
item.accessRightsOpen Access-
item.fulltextWith Fulltext-
item.contributorVamplew, Peter-
item.contributorSmith, Benjamin J.-
item.contributorKallstrom, Johan-
item.contributorRamos, Gabriel-
item.contributorRadulescu, Roxana-
item.contributorRoijers, Diederik M.-
item.contributorHayes, Conor F.-
item.contributorHeintz, Fredrik-
item.contributorMannion, Patrick-
item.contributorLIBIN, Pieter-
item.contributorDazeley, Richard-
item.contributorFoale, Cameron-
item.validationecoom 2023-
crisitem.journal.issn1387-2532-
crisitem.journal.eissn1573-7454-
Appears in Collections:Research publications
Files in This Item:
File Description SizeFormat 
Scalar reward is not enough_ a response to Silver, Singh, Precup and Sutton (2021).pdfPublished version645.59 kBAdobe PDFView/Open
Show simple item record

WEB OF SCIENCETM
Citations

8
checked on Oct 4, 2024

Google ScholarTM

Check

Altmetric


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.