AAAI Publications, Twenty-Third International FLAIRS Conference

Font Size: 
A Reinforcement Learning Model for Economic Agent Specialization
Denton Cockburn, Ziad Kobti, Timothy A. Kohler

Last modified: 2010-05-05


We model a framework whereby agents decide how to allocate their time among available tasks. Agents learn from their previous experiences, adjusting the weights given to each task as a result. These social agents are also influenced by the experiences of others in their social networks, including kin and trading partners. Agents are allowed to trade surplus goods and to request goods that they need using various trading methods. We demonstrate this model using the Agent-Based Model of the Village Ecodynamics Project, which simulates the life of Pueblo farmers of the central Mesa Verde region between A.D. 600 and 1300.


agent specialization; multi-agent simulation; social agents

Full Text: PDF