Gossip, Sexual Recombination and the El Farol Bar: Modelling the Emergence of Heterogeneity

Abstract

Brian Arthur's 'El Farol Bar' model is extended so that the agents also learn and communicate. The learning and communication is implemented using an evolutionary process acting upon a population of mental models inside each agent. The evolutionary process is based on a Genetic Programming algorithm. Each gene is composed of two tree-structures: one to control its action and one to determine its communication.A detailed case-study from the simulations show how the agents have differentiated so that by the end of the run they had taken on very different roles. Thus the introduction of a flexible learning process and an expressive internal representation has allowed the emergence of heterogeneity.

Coordination in the El Farol Bar problem: The role of social preferences and social networks

“…Briefly put, there are two kinds of learning mechanism being studied in the literature. The first one is best-response learning (Arthur, 1994;Edmonds, 1999;Fogel et al, 1999;Challet et al, 2004;Atilgan, Atilgan and Demirel, 2008) and the second one is reinforcement learning (Bell and Sethares, 1999;Franke, 2003;Zambrano, 2004;Whitehead, 2008). 1 The typical results are as follows.…”

“…In the first line of development, Edmonds (1999) proposes an extension of the El Farol Bar problem where agents can change their strategies set by means of genetic programming and are given the chance to communicate with other agents before making their decision as to whether to go to the El Farol Bar. Simulations show that, as in the original model, the attendance at the bar fluctuates around the threshold level, and does not seem to settle down into any regular pattern.…”

Building coalitions involving agents and humans

“…Drawing on past work [2], we extended the El Farol bar model in two directions. We allowed agents to communicate by exchanging private messages.…”

Complex Science for a Complex World : Exploring Human Ecosystems with Agents

“…Source: Arthur, 1994 The bar problem extended Given that agents do communicate with each other in the real world, we wonder what may happen if they are permitted to exchange information? Bruce Edmonds' work allows communication among agents before they make their final decisions whether or not to go to the bar (Edmonds 1999). Using a genetic programming algorithm to simulate adaptive learning, he allows each agent to competitively develop its models of what the other agents are going to do.…”