Adversarial vulnerabilities of human decision-making.
decision-making
recurrent neural networks
reinforcement learning
Journal
Proceedings of the National Academy of Sciences of the United States of America
ISSN: 1091-6490
Titre abrégé: Proc Natl Acad Sci U S A
Pays: United States
ID NLM: 7505876
Informations de publication
Date de publication:
17 11 2020
17 11 2020
Historique:
pubmed:
6
11
2020
medline:
26
1
2021
entrez:
5
11
2020
Statut:
ppublish
Résumé
Adversarial examples are carefully crafted input patterns that are surprisingly poorly classified by artificial and/or natural neural networks. Here we examine adversarial vulnerabilities in the processes responsible for learning and choice in humans. Building upon recent recurrent neural network models of choice processes, we propose a general framework for generating adversarial opponents that can shape the choices of individuals in particular decision-making tasks toward the behavioral patterns desired by the adversary. We show the efficacy of the framework through three experiments involving action selection, response inhibition, and social decision-making. We further investigate the strategy used by the adversary in order to gain insights into the vulnerabilities of human choice. The framework may find applications across behavioral sciences in helping detect and avoid flawed choice.
Identifiants
pubmed: 33148802
pii: 2016921117
doi: 10.1073/pnas.2016921117
pmc: PMC7682379
doi:
Types de publication
Journal Article
Research Support, Non-U.S. Gov't
Langues
eng
Sous-ensembles de citation
IM
Pagination
29221-29228Informations de copyright
Copyright © 2020 the Author(s). Published by PNAS.
Déclaration de conflit d'intérêts
The authors declare no competing interest.
Références
PLoS Comput Biol. 2015 Jun 08;11(6):e1004254
pubmed: 26053429
Nat Commun. 2019 Jun 26;10(1):2808
pubmed: 31243285
Science. 1974 Sep 27;185(4157):1124-31
pubmed: 17835457
Nature. 2015 Feb 26;518(7540):529-33
pubmed: 25719670
Nat Commun. 2019 May 24;10(1):2319
pubmed: 31127115
Science. 2005 Apr 1;308(5718):78-83
pubmed: 15802598
PLoS Comput Biol. 2019 Jun 11;15(6):e1006903
pubmed: 31185008