Cooperation Emergence under Resource-Constrained Peer Punishment

By Samhar Mahmoud, Simon Miles, Michael Luck. 2016

19 citations recorded by the source corpus.

In distributed computational systems with no central authority, social norms have shown great potential in regulating the behaviour of self-interested agents, due to their distributed cost. In this context, peer punishment has been an important instrument in enabling social norms to emerge, and such punishment is usually assigned a certain enforcement cost that is paid by agents applying it. However, models that investigate the use of punishment as a mechanism to allow social norms to emerge usually assume that unlimited resources are available to agents to cope with the resulting enforcement costs, yet this assumption may not hold in real world computational systems, since resources are typically limited and thus need to be used optimally. In this paper, we use a modified version of the metanorm model originally proposed by Axelrod to investigate this, and show that it allows norm emergence only in limited cases under bounded resources. In response, we propose a resource-aware adaptive punishment technique to address this limitation, and give an experimental evaluation of the new technique that shows it enables norm establishment under limited resources.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Multi-agent Reinforcement Learning in Sequential Social Dilemmas (2017, background): “Our main result is that both of the SSDs we considered, Gathering and Wolfpack, have empirical payoﬀ matrices that are Prisoner’s Dilemma (PD).”