We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Ancillary-file links:

Ancillary files (details):

Current browse context:

q-bio

Change to browse by:

References & Citations

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Quantitative Biology > Populations and Evolution

Title: Evolutionary instability of selfish learning in repeated games

Abstract: Across many domains of interaction, both natural and artificial, individuals use past experience to shape future behaviors. The results of such learning processes depend on what individuals wish to maximize. A natural objective is one's own success. However, when two such "selfish" learners interact with each other, the outcome can be detrimental to both, especially when there are conflicts of interest. Here, we explore how a learner can align incentives with a selfish opponent. Moreover, we consider the dynamics that arise when learning rules themselves are subject to evolutionary pressure. By combining extensive simulations and analytical techniques, we demonstrate that selfish learning is unstable in most classical two-player repeated games. If evolution operates on the level of long-run payoffs, selection instead favors learning rules that incorporate social (other-regarding) preferences. To further corroborate these results, we analyze data from a repeated prisoner's dilemma experiment. We find that selfish learning is insufficient to explain human behavior when there is a trade-off between payoff maximization and fairness.
Comments: 73 pages; final version
Subjects: Populations and Evolution (q-bio.PE)
DOI: 10.1093/pnasnexus/pgac141
Cite as: arXiv:2105.06199 [q-bio.PE]
  (or arXiv:2105.06199v3 [q-bio.PE] for this version)

Submission history

From: Alex McAvoy [view email]
[v1] Thu, 13 May 2021 11:37:09 GMT (9710kb,D)
[v2] Mon, 30 Aug 2021 15:22:19 GMT (9773kb,D)
[v3] Wed, 31 Aug 2022 23:24:53 GMT (10996kb,AD)

Link back to: arXiv, form interface, contact.