Conference Proceedings

Adaptive Risk Aversion in Social Dilemmas

Michael Kirley, Friedrich Burkhard von der Osten, G Dick (ed.), WN Browne (ed.), P Whigham (ed.), M Zhang (ed.), LT Bui (ed.), H Ishibuchi (ed.), Y Jin (ed.), X Li (ed.), Y Shi (ed.), P Singh (ed.), KC Tan (ed.), K Tang (ed.)

Lecture Notes in Artificial Intelligence | SPRINGER-VERLAG BERLIN | Published : 2014

Abstract

Risks are ubiquitous in many social dilemmas. A decision to cooperate (or invest in a common pool) is inherently “risky” when the reward received depends on the frequency of actions displayed by other social group members. In this paper, we study the evolutionary dynamics of the multiple-group public goods game where the population of mobile individuals is divided into fixed-sized groups. Agents use continuous investment strategies. This approach allows for the evolution of an intermediate level of investments correlated with specific risk aversion levels. Agents are also presented with an opportunity to switch groups at the end of each round based on an environmental trigger. Detailed simul..

View full abstract