Network-Iterated Prisoner's Dilemma

TL;DR

This paper introduces a stochastic network model of the Iterated Prisoner's Dilemma where players adapt strategies locally, revealing a new highly successful strategy and providing theoretical metrics for strategy success.

Contribution

The paper presents a novel stochastic network model for the Iterated Prisoner's Dilemma with local strategy updates and identifies a new superior strategy.

Findings

A new highly effective strategy naturally emerges in the model.

The model demonstrates the flexibility and noise tolerance of strategies.

Theoretical parameters are derived to gauge strategy success.

Abstract

We introduce the stochastic Network-Iterated Prisoner's Dilemma (NIPD) model, a network of players playing the Prisoner's Dilemma with their neighbours, each with a memory-one strategy which they constantly and locally update to improve their success. This process is non-deterministic, and mirrors societal interactions in many relevant aspects. We use it to assess the flexibility, noise tolerance and real-world adaptability of some well-known strategies. Furthermore, in the model a new strategy naturally emerges which proves way more successful than those. We also derive some theoretical parameters that gauge the success of a strategy in this context.