|
Vol: 55(69) No: 4 / December 2010 A Comparative Study of Various Evolutionary Algorithms and Their Combinations for Optimizing Fuzzy Rule based Inference Systems Zsolt Dányádi Department of Telecommunications and Media Informatics, Budapest University of Technology and Economics, Budapest, Hungary, e-mail: dz602@hszk.bme.hu Krisztián Balázs Department of Telecommunications and Media Informatics, Budapest University of Technology and Economics , Budapest, Hungary, e-mail: balazs@tmit.bme.hu László T. Kóczy Institute of Informatics, Electrical and Mechanical Engineering, Faculty of Engineering Sciences, Széchenyi István University, Győr, Hungary, e-mail: koczy@sze.hu Keywords: evolutionary algorithm, fuzzy inference, rule-base optimization, co-evolution, fuzzy rule interpolation, multiple populations Abstract The goal of this paper is to provide an overview of a variety of evolutionary algorithms, combined with co-evolutionary extensions, and compare their efficiency when applied to fuzzy rule-based inference and learning. Four basic evolutionary methods are presented in this paper, each working with variations on the basic principles, along with two types of auxiliary populations, which provide new mechanisms for generating new individuals. A method that uses multiple populations with different algorithms and connects them through migrations was also included in the comparisons. Fuzzy rule-based inference can be used to model an appropriate outward behavior of a system when given a specific input, which, in the case of this comparative study, is determined by a set of samples. The samples were generated from sufficiently complex objective functions, as well as existing databases, depending on the problem. Optimizing a fuzzy rule-based inference system is a matter of finding a rule-base that is as close to producing the desired behavior as possible. Therefore, the evolutionary algorithms presented here work on rule-base descriptors as the genetic makeup of individuals. These algorithms were applied to real and categorical decision problems, using two different kinds of inference. The results were then compared to determine the general relations between the efficiency of the various evolutionary methods. References [1] K. Balázs, L. T. Kóczy, J. Botzheim, "Comparison of Fuzzy Rule-based Learning and Inference Systems", Proceedings of the 9th International Symposium of Hungarian Researchers on Computational Intelligence and Informatics, CINTI 2008, Budapest, Hungary, 2008, pp. 61-75. [2] J. H. Holland, "Adaption in Natural and Artificial Systems", The MIT Press, Cambridge, Massachusetts, 1992. [3] N. E. Nawa and T. Furuhashi, “Fuzzy system parameters discovery by bacterial evolutionary algorithm”, IEEE Transactions on Fuzzy Systems, 7(5), 1999, pp. 608-616. [4] S. H. Jung, “Queen-bee evolution for genetic algorithms”, Electronic Letters, 20 Mar. 2003, Vol. 39, No. 6, pp. 575-576. [5] M. F. Azeem, Alam M. Saad : “Modified Queen Bee Evolution Based Genetic Algorithm for Tuning of Scaling Factors of Fuzzy Knowledge Base Controller”, India Annual Conference 20-22 Dec. 2004, 5 pp. [6] J. Brownlee, “Clonal Selection Algorithms”, CIS Technical Report 070209A, Feb. 2007. [7] N. Kubota, T. Fukuda, T. Arakawa, K. Shimojima : “Evolutionary Transition on Virus-Evolutionary Genetic Algorithm”, IEEE International Conference on Evolutionary Computation, 1997, pp. 291-296. [8] Naoyuki Kubota, Toshio Fukuda, Takemasa Arakawa, Koji Shimojima, “Virus-Evolutionary Algorithm with Subpopulations”, IEEE International Conference on Systems, Man, and Cybernetics, 1996, Vol. 3, pp. 1930-1935. [9] D. Wallin, C. Ryan, R. M. A. Azad, “Symbiogenetic Co-evolution” – Congress on Evolutionary Computation 2005, pp. 1613-1620. [10] R. A. Watson, J. B. Pollack, “How Symbiosis Can Guide Evolution” - Lecture Notes In Computer Science; Vol. 1674, 1990, 10 [11] A. Deepak, P. Samuel, “An evolutionary multi population approach for test data generation”, World Congress on Nature & Biologically Inspired Computing, 2009. NaBIC 2009, 9-11 Dec. 2009, pp. 1451-1456. [12] Bi Li, Tu-Sheng Lin, Liang Liao, Ce Fan, “Genetic algorithm based on multipopulation competitive co-evolution”, Proceedings of IEEE Congress on Evolutionary Computation, CEC 2008, 2008, pp. 225-228. [13] A. E. Ruano, C. Cabrita, J. V. Oliveira, and L. T. Kóczy. Supervised training algorithms for B-spline neural networks and neuro-fuzzy systems. International Journal of Systems Science, 33(8):689-711, 2002. [14] N. E. Nawa and T. Furuhashi. Fuzzy system parameters discovery by bacterial evolutionary algorithm. IEEE Transactions on Fuzzy Systems, 7(5):608-616, Oct. 1999. [15] K. Balázs, J. Botzheim, L. T. Kóczy, “Comparative Analysis of Various Evolutionary and Memetic Algorithms”, 10th International Symposium of Hungarian Researchers on Computational Intelligence and Informatics, pp. 193-205. [16] K. Balázs, J. Botzheim, L. T. Kóczy: Comparison of Various Evolutionary and Memetic Algorithms, Proceedings of the International Symposium on Integrated Uncertainty Management and Applications, IUM 2010, Ishikawa, Japan, 2010, pp. 431-442. [17] K. Balázs, J. Botzheim, L. T. Kóczy: Comparative Analysis of Interpolative and Non-interpolative Fuzzy Rule Based Machine Learning Systems Applying Various Numerical Optimization Methods, World Congress on Computational Intelligence, WCCI 2010, Barcelona, Spain, 2010, pp. 875-982. [18] Zs. Dányádi, K. Balázs, L. T. Kóczy: A Comparative Study of Various Evolutionary Algorithms Used for Fuzzy Rule-based Inference and Learning Systems, IEEE International Joint Conferences on Computational Cybernetics and Technical Informatics, ICCC-CONTI 2010, Timisoara, Romania, 2010, pp. 49-54. [19] L. T. Kóczy, K. Hirota, “Approximate Reasoning by Linear Rule Interpolation and General Approximation”, Internat. J. Approx. Reason., 9, 1993, pp. 197-225. [20] D. Tikk, I. Joó, L. T. Kóczy, P. Várlaki, B. Moser and T. D. Gedeon: Stability of interpolative fuzzy KH-controllers, Fuzzy Sets and Systems, Vol. 125, pp. 105-119, 2002. [21] E. H. Mamdani, “Application of Fuzzy Algorithms for Control of Simple Dynamic Plant”, Proceedings of the IEEE, Vol. 121, No. 12, pp. 1585-1588, 1974. [22] A. Frank and A. Asuncion: UCI Machine Learning Repository [http://archive.ics.uci.edu/ml]. Irvine, CA: University of California, School of Information and Computer Science, 2010. |