Evolving an ecology of two-tiered organizations

Evolutionary models typically rely on a single level of evolution for training a team of cooperating agents. I present a model that evolves at two levels—an “organizational” level and the more traditional “individual” level. Each organization contains an embedded agent population that goes through a full evolutionary process every organizational time-step. The organization’s genetic code is essentially a policy that specifies the training process for its embedded agents. It also defines the creation of a representative team that is compiled after each organizational time-step. An organization’s fitness is based on the performance of this representative team. Categories and Subject Descriptors I.2.8-Problem Solving, Control Methods and Search General Terms Algorithms, Design Keywords Evolutionary computation, genetic programming, genetic algorithms, distributed artificial intelligence, teamwork

• Type: Conference
• Year: 2005
• Where: GECCO
• Authors: Travis Kriplean
• Keywords: Embedded Agents, GECCO 2005, Organizational Time-step, Representative Team