Fuzzy cultural algorithms with evolutionary programming
Cultural Algorithms represent computational models based upon principles of cultural evolution . They consist of a population component, a belief space containing knowledge compiled from individual experience, and a communication protocol which controls the interaction between two components. This protocol consists of an acceptance function and an influence function. The acceptance function determines how individuals can affect the knowledge contained in the belief space. The influence function determines how that knowledge can affect the individuals in the population.
The communication protocols used in early versions of Cultural Algorithms were inherently deterministic in nature. Recently, Chung  developed a fuzzy acceptance function based upon linear membership functions for use in real-valued function optimization. This system used an evolutionary programming population model and generated significant improvements overall when compared to the same system with a deterministic acceptance function. But, there was a subset of problems for which there was still ample room for improvement.
Here, we employ continuous non-linear membership functions to control the acceptance of individual contributions to the belief space. The approach is used with the Cultured EP system in order to solve 12 problems taken from Chung's original test suite. The results, based upon 25 trails for each function, indicated a statistically significant (0.05 level) improvement in performance overall for these 12 functions.
Unable to display preview. Download preview PDF.
- Robert G. Reynolds and Bernard P. Zeigler, “A Formal Mathematical Model for the Operation of Consensus-Based Hunting-Gathering Bands, Mathematical Approaches to Culture Change, Academic Press, Inc., 1979Google Scholar
- Chan-Jin Chung and Robert G. Reynolds, “A Testbed for Solving Optimization Problems Using Cultural Algorithms,” in Proceedings of the Fifth Annual Conference on Evolutionary Programming, 1996Google Scholar
- Chan-Jin Chung “Knowledge-based Approach to Self Adoption in Cultural Algorithms” Ph. D. Dissertation in Computer Science Dept. at Wayne State University, 1996.Google Scholar
- Peter A. Angeline, “Adaptive and Self-Adaptive Evolutionary Computation,” in Computation Intelligence, IEEE Press, New York, 1995, pp. 152–163.Google Scholar
- Thomas Back and Hans-Paul Schwefel, “Evolutionary Computation: Overview,” in Proceedings of IEEE Intentional Conference Evolutionary Computation (ICEC '96), Nagoya, Japan pp. 20–29.Google Scholar
- Zbigniew Michalewicz, Genetic Algorithms + Data Structures Evolution Program, (Second, Extended Edition,) Springer-Verlag, Berlin Heidelberg New York, 1994Google Scholar
- B. P. Buckles and F. E. Perty. “Fuzzy sets and their applications” In: Gupta and Sanchez pp. 377–387, 1970Google Scholar
- George J. Klir, Bo Yuan “Fuzzy Sets and Fuzzy Logic” Prentice Hall PTR, New Jersey, 1995Google Scholar
- D Dubois, H. Prade “Vagueness, typicality and uncertainty in class hierarchies” Int. Journal of Intelligent Systems 6(6)167–183, 1991Google Scholar
- David B. Fogel, “Using Fitness Distributions to Design More Efficient Evolutionary Computations” in Proceedings of IEEE International Conference on Evolutionary' Computation (ICEC, 96), pp. 11–19.Google Scholar
- H. P. Schwefel, Numerical Optimization of Computer Models John Wiley, Chichester, UK, 1981Google Scholar
- Xin Yao and Yong Liu, “Fast Evolutionary Programming,” in Proceedings of the Fifth Annual Conference on Evolutionary Programming, 1996Google Scholar
- Eu, Yieqin and Zhang, Xiandie “ Fuzzy Sets and Their Applications” Chendu Telecommunication Engineering College Published in 1991 [Chinese]Google Scholar
- L. A. Zadeh “Fuzzy Logic for The management of Uncertainty” John Wisley, New York, 1992Google Scholar
- Zhu, Shinin. “Apply the fuzzy Theory to Chinese Traditional Medical Diagnosis” (Chinese) 5th National Conference on “Fuzzy Set and Application”. May. 1986. Shanghai, China. pp. 331–339.Google Scholar
- Zhu, Shinin. “Apply the Nearest Rule to Object Oriented Database” IEEE International Conference on Database and Expert Systems Applications. Sept. 1996. Zurich, Switzerland. pp. 482–491.Google Scholar