Cooperative interactive cultural algorithms adopting knowledge migration

ABSTRACT

In many optimization problems with implicit indexes, human need to participate in the evaluation process synchronously in different computer nodes. And human is easy to feel tired. In order to alleviate human fatigue, implicit knowledge embodied in the evolution process, which reflect human cognition and preference, is extracted and utilized. However, how to effectively exchange information among nodes is not taken into account. Aiming at systemic analysis and effective application about implicit knowledge, cooperative interactive cultural algorithm adopting knowledge migration strategy is proposed. A novel knowledge model based on characteristic-vector is adopted to describe implicit knowledge embodied in the evolution process, including human cognitive tendency, the degree of human preference, the degree of human fatigue and human cognition schema. According to the evolution status of population and human fatigue in each computer node, human cognition schemas are migrated between nodes. And common knowledge is obtained by coordination strategy and utilized to induce the evolution process of ICA in each computer node. Taking cooperative fashion design system as a testing platform, the rationality of knowledge migration strategy is proved. Simulation results indicate this algorithm can alleviate human fatigue and improve the speed of convergence effectively.

REFERENCES

• 1: Takagi H.Interactive Evolutionary Computation:Fusions of The Capabilities of EC Optimization and Human Evaluation, Proc. of the IEEE CEC, (2001) 1275--1296
• 2: Guo Yi-nan, Gong Dun-wei. Interactive genetic algorithms based on implicit knowledge model.Lecture Notes in Computer Science.(2006) 4247: 369--376
• 3: B.Peng.Knowledge and population swarms in cultural algorithms for dynamic environments. USA: Wayne State University.(2005)
• 4: Gong D., Zhou Y., and Li T., Cooperative interactive genetic algorithm based on user's preference, International Journal of Information Technology.(2005) 11(10): 1--10.
• 5: R.P.Wiegand,W.C.Liles,K.A.De Jong.An empirical analysis of collaboration methods in cooperative co-evolutionary algorithm.Proceedings of the Genetic and Evolutionary Computation Conference.(2001) 1235-- 1245
• 6: Guo Yi-nan, Cheng Jian, Gong Dun-wei,et al. Knowledge-inducing interactive genetic algorithms based on multi-agent.Lecture Notes in Computer Science. (2006) 4221: 759-- 768.
• 7: Kim H., Cho S.. Application of interactive genetic algorithm to fashion design. Engineering Applications of Artificial Intelligence. (2000)13(6): 635--644.

INDEX TERMS

Primary Classification:
  G. Mathematics of Computing
  G.1 NUMERICAL ANALYSIS
      G.1.6 Optimization
          Subjects: Global optimization

General Terms:
Algorithms

Keywords:
coordination, implicit knowledge, interactive cultural algorithms, knowledge migration