Surviving the Tragedy of Commons: Emergence of Altruism in a Population of Evolving Autonomous Agents

Jean-Marc Montanier 1, 2 Nicolas Bredeche 1, 2
1 TAO - Machine Learning and Optimisation
CNRS - Centre National de la Recherche Scientifique : UMR8623, Inria Saclay - Ile de France, UP11 - Université Paris-Sud - Paris 11, LRI - Laboratoire de Recherche en Informatique
Abstract : This paper explores the following question: how a fixed-size population of autonomous agents (such as a swarm of robotic agents) may evolve altruistic behaviors during open-ended evolution. In particular, we focus on a situation where the tragedy of commons can possibly occur: a situation where individuals must display altruistic behaviors in order for the whole population to avoid extinction. Our approach considers a sub-individual framework, defined at the level of genomes rather than agents, in order to provide an efficient algorithmic solution for the emergent of coordination among the population. Experiments show that the proposed evolutionary adaptation algorithm favors the emergence of altruistic behavior under some assumptions regarding genome relatedness. In-depth experimental studies explore the relation between genotypic diversity and degree of altruism as well as the exact nature of the evolutionary adaptation process.
Document type :
Conference papers
Complete list of metadatas

Cited literature [34 references]  Display  Hide  Download

https://hal.inria.fr/inria-00601776
Contributor : Nicolas Bredeche <>
Submitted on : Monday, June 20, 2011 - 2:40:50 PM
Last modification on : Thursday, April 5, 2018 - 12:30:12 PM
Long-term archiving on : Friday, November 9, 2012 - 4:35:18 PM

File

2011-ECAL-emergenceofaltruism....
Files produced by the author(s)

Identifiers

  • HAL Id : inria-00601776, version 1

Collections

Citation

Jean-Marc Montanier, Nicolas Bredeche. Surviving the Tragedy of Commons: Emergence of Altruism in a Population of Evolving Autonomous Agents. European Conference on Artificial Life, Aug 2011, Paris, France. ⟨inria-00601776⟩

Share

Metrics

Record views

334

Files downloads

253