Using GA for evolving weights in neural networks

• Autorzy: Hameed Wafaa Mustafa , Kanbar Asan Baker

Identyfikatory

DOI

10.23743/acs-2019-18

• Języki publikacji: EN

Abstrakty

EN

This article aims at studying the behavior of different types of crossover operators in the performance of Genetic Algorithm. We have also studied the effects of the parameters and variables (crossover probability (Pc), mutation probability (Pm), population size (popsize) and number of generation (NG) for controlling the algorithm. This research accumulated most of the types of crossover operators these types are implemented on evolving weights of Neural Network problem. The article investigates the role of crossover in GAs with respect to this problem, by using a comparative study between the iteration results obtained from changing the parameters values (crossover probability, mutation rate, population size and number of generation). From the experimental results, the best parameters values for the Evolving Weights of XOR-NN problem are NG = 1000, popsize = 50, Pm = 0.001, Pc = 0.5 and the best operator is Line Recombination crossover.

Słowa kluczowe

EN

genetic algorithm; neural network; crossover; mutation

PL

algorytm genetyczny; sieć neuronowa; skrzyżowanie; mutacja

• Wydawca: Polskie Towarzystwo Promocji Wiedzy ; Lublin University of Technology
• Czasopismo: Applied Computer Science
• Rocznik: 2019
• Tom: Vol. 15, no 3
• Strony: 21--33
• Opis fizyczny: Bibliogr. 13 poz., fig., tab.

Twórcy

autor

Hameed Wafaa Mustafa

• Cihan University, Department of Computer Science, Sulaimaniya, Iraq

autor

Kanbar Asan Baker

• Cihan University, Department of Computer Science, Sulaimaniya, Iraq

Bibliografia

• [1] Al-Inazy, Q. A. (2005). A Comparison between Lamarckian Evolution and Behavior Evolution of Neural Network (Unpublished M.Sc. Thesis). Al- Mustansriyah University, Baghdad, Iraq.
• [2] Arjona, D. (1996). A hybrid artificial neural network/genetic algorithm approach to on-line operations for the optimization of electrical power systems. In IECEC 96. Proceedings of the 31st Intersociety Energy Conversion Engineering Conference (pp. 2286–2290 vol. 4). Washington, DC, USA. doi:10.1109/IECEC.1996.561174
• [3] Goldberg, D. E. (1989). Genetic Algorithms in search, Optimization, and Machine Learning. Boston, MA, USA: Addison–Wesley Longman Publishing Co., Inc.
• [4] Koza, J. R. (1992). Genetic programming: on the programming of computers by means of natural selection. Cambridge, MA, USA: MIT Press.
• [5] Michalewicz, Z. (1996). Genetic Algorithm + Data Structure = Evolution Programs, 3rd Revised Extended Edition. New York, USA: Springer – Verlag Berlin Heidelberg.
• [6] Mitchell, M. (1998). An Introduction of Genetic Algorithms. Cambridge, MA, USA: MIT Press.
• [7] Montana, D., & Davis, L. (1989). Training Feed Forward neural networks using Genetic Algorithms, In IJCAI'89 Proceedings of the 11th international joint conference on Artificial intelligence (pp. 762–767). San Francisco, CA, USA: Morgan Kaufmann Publishers Inc.
• [8] Hameed, W. M., & Kanbar, A. B. (2017). A Comparative Study of Crossover Operators for Genetic Algorithms to Solve Travelling Salesman Problem. International Journal of Research – Granthaalayah, 5(2), 284–291. doi:10.5281/zenodo.345734
• [9] Hameed, W. M. (2016). The Role of Crossover on Optimization of a Function Problem Using Genetic Algorithms. International Journal of Computer Science and Mobile Computing, 5(7), 425–429.
• [10] Weisman, O., & Pollack, Z. (2002). Neural Networks Using Genetic Algorithm. Retrieved from http://www.cs.bgu.ac.il/NNUGA.
• [11] Whitley, D., Starkweather, T., & Fuquay, D. A. (1989). Scheduling Problems and Traveling Salesman: The Genetic Edge Recombination Operator. ICGA.
• [12] Whitley, D. (1995). Genetic Algorithms and Neural Networks. In J. Periaux & G. Winter (Eds.), Genetic Algorithms in Engineering and Computer Science (pp. 191-201). John Wiley & Son Corp.
• [13] Wright, A. H. (1991). Genetic Algorithms for Real Parameters Optimization. Foundation of Genetic Algorithms, 1, 205-218. doi:10.1016/B978-0-08-050684-5.50016-1

Uwagi

Opracowanie rekordu w ramach umowy 509/P-DUN/2018 ze środków MNiSW przeznaczonych na działalność upowszechniającą naukę (2019).