Exploring knowledge and population swarms via an agent-based Cultural Algorithms Simulation Toolkit (CAT)

Cultural algorithms employ a basic set of knowledge sources, each related to knowledge observed in various social species. These knowledge sources are then combined to direct the decisions of the individual agents in solving optimization problems. While many successful real- world applications of cultural algorithms have been produced, we are interested in studying the fundamental computational processes involved in the use of cultural systems as problem solvers. Here we describe a Java-based toolkit system, the cultural algorithm toolkit (CAT) developed in the repast symphony simulation environment. The system allows users to easily configure and visualize the problem solving process of a cultural algorithm. Currently the system supports predator/prey problem solving in a "cones world" environment as well as a suite of benchmark problems in engineering design. Example runs of a predator prey example are presented to demonstrate the systems\´ capabilities.