Knowledge Learning and Social Swarms in Cultural Systems

ABSTRACT In previous work it was observed that certain problem-solving phases emerged during the optimization process for a real-valued functional surface formed within a cones-world environment using a Cultural Algorithm. The Cultural Algorithm was configured using five knowledge sources in the belief space, and an Evolutionary Programming model for the population space. It turned out that the five knowledge sources exhibited a swarming behavior at the meta-level while solving the problem (Iacoban et al., 2003). In this paper we investigate whether this swarming behavior at the meta-level induces swarming at the population level. Our results suggest that the interaction or swarming of knowledge sources at the meta-level induce a swarming of individuals at the population level. That is, collective or flocking behavior rather than being initally intentional can be produced implicitly as a result of individuals using the same information to make their own individual decisions.

[1]  R.W. Morrison,et al.  A test problem generator for non-stationary environments , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).

[2]  R. Reynolds,et al.  Using knowledge-based evolutionary computation to solve nonlinear constraint optimization problems: a cultural algorithm approach , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).

[3]  Robert G. Reynolds,et al.  Knowledge-based solution to dynamic optimization problems using cultural algorithms , 2001 .

[4]  Robert G. Reynolds,et al.  Cultural swarms: modeling the impact of culture on social interaction and problem solving , 2003, Proceedings of the 2003 IEEE Swarm Intelligence Symposium. SIS'03 (Cat. No.03EX706).

[5]  Robert G. Reynolds,et al.  CAEP: An Evolution-Based Tool for Real-Valued Function Optimization Using Cultural Algorithms , 1998, Int. J. Artif. Intell. Tools.