Scaffolding cooperation in human groups with deep reinforcement learning.

Autor: McKee KR; Google DeepMind, London, UK. kevinrmckee@google.com., Tacchetti A; Google DeepMind, London, UK., Bakker MA; Google DeepMind, London, UK., Balaguer J; Google DeepMind, London, UK., Campbell-Gillingham L; Google DeepMind, London, UK., Everett R; Google DeepMind, London, UK., Botvinick M; Google DeepMind, London, UK.; Gatsby Computational Neuroscience Unit, University College London, London, UK.
Jazyk: angličtina
Zdroj: Nature human behaviour [Nat Hum Behav] 2023 Oct; Vol. 7 (10), pp. 1787-1796. Date of Electronic Publication: 2023 Sep 07.
DOI: 10.1038/s41562-023-01686-7
Abstrakt: Effective approaches to encouraging group cooperation are still an open challenge. Here we apply recent advances in deep learning to structure networks of human participants playing a group cooperation game. We leverage deep reinforcement learning and simulation methods to train a 'social planner' capable of making recommendations to create or break connections between group members. The strategy that it develops succeeds at encouraging pro-sociality in networks of human participants (N = 208 participants in 13 groups) playing for real monetary stakes. Under the social planner, groups finished the game with an average cooperation rate of 77.7%, compared with 42.8% in static networks (N = 176 in 11 groups). In contrast to prior strategies that separate defectors from cooperators (tested here with N = 384 in 24 groups), the social planner learns to take a conciliatory approach to defectors, encouraging them to act pro-socially by moving them to small highly cooperative neighbourhoods.
(© 2023. The Author(s).)
Databáze: MEDLINE
Externí odkaz: