الگوریتم رقابت استعماری آشوبی متعامد اصلاح شده و بکارگیری آن در بهبود بازشناسی الگو در شبکۀ عصبی پرسپترون¬های چند لایه
محورهای موضوعی : عمومىپیمان معلم 1 , مهرداد صادقی حریری 2 , مهدی هاشمی 3
1 - هیات علمی گروه مهندسی برق
2 - دانشگاه آزاد اسلامی واحد علوم و تحقیقات تهران
3 - موسسه آموزش عالی پیام گلپایگان، دانشکده مهندسی برق
کلید واژه: الگوریتم رقابت استعماری آشوبی متعامد, شبکه عصبی پرسپترون چند لایه, طبقه¬بندی داده,
چکیده مقاله :
علی رغم موفقیت الگوریتم رقابت استعماری (ICA) در حل مسائل بهینه سازی، این الگوریتم کماکان از به دام افتادن مکرر در کمینه محلی و سرعت پایین همگرایی رنج می برد. در این مقاله، نسخۀ جدیدی از این الگوریتم، به نام رقابت استعماری آشوبی متعامد اصلاح شده (COICA)، پیشنهاد می شود. در سیاست جذب نسخه پیشنهادی، هرمستعمره از طریق تعریف بردار متعامد نوینی، فضای حرکت به سمت استعمارگر را جستجو می کند. همچنین احتمال انتخاب امپراطوری های قدرتمند، از طریق تابع توزیع بولتزمان تعریف شده و عمل انتخاب از طریق روش چرخ رولت انجام گرفته است. از الگوریتم پیشنهادی برای آموزش شبکه عصبی پرسپترون چند لایه (MLP) جهت طبقه بندی مجموعه داده های استاندارد، از جمله یونسفر و سونار استفاده شده است. برای ارزیابی عملکرد این الگوریتم و بررسی میزان تعمیم پذیری شبکه عصبی آموزش ديده با نسخه پيشنهادی، از روش اعتبارسنجی متقابل K-Fold استفاده شده است. نتایج بدست آمده از شبیه سازی ها، کاهش خطای آموزش شبکه و همچنین بهبود تعمیم پذیری الگوریتم پیشنهادی را تایید می کند.
Despite the success of the Colonial Competition Algorithm (ICA) in solving optimization problems, this algorithm still suffers from repeated entrapment in the local minimum and low convergence speed. In this paper, a new version of this algorithm, called Modified Orthogonal Chaotic Colonial Competition (COICA), is proposed. In the policy of absorbing the proposed version, each colony seeks the space to move towards the colonizer through the definition of a new orthogonal vector. Also, the possibility of selecting powerful empires is defined through the boltzmann distribution function, and the selection operation is performed through the roulette wheel method. The proposed multilevel perceptron neural network (MLP) algorithm is used to classify standard datasets, including ionosphere and sonar. To evaluate the performance of this algorithm and to evaluate the generalizability of the trained neural network with the proposed version, the K-Fold cross-validation method has been used. The results obtained from the simulations confirm the reduction of network training error as well as the improved generalizability of the proposed algorithm.
1.M. Melanie, “An Introduction to Genetic Algorithms,” Massachusett's MIT Press, 34(7):pp.1-9, 1999.
2.L. A. Ingber, “Simulated annealing: practice versus theory,” J. Math. Comput. Modell., 18(11):pp. 29–57, 1993.
3.Kennedy, J., Eberhart, R.C.: Particle Swarm Optimization. Proceedings of IEEE, 1942– 1948 (1995)
4.M. Dorigo, V. Maniezzo and A. Colorni, “The ant system: optimization by a colony of cooperating agents,” IEEE Transaction System Man Cybern, B 26(1):pp. 29–41, 1996.
5.B. Franklin and M. Bergerman, “Cultural Algorithms: Concepts and Experiments,” In Proceedings of the IEEE Congress on Evolutionary Computation, 2: pp. 1245–1251, 2000.
6.R. Storn and K. Price, “Differential evolution - a simple and efficient heuristic for global optimization over continuous spaces,” Journal of Global Optimization, 11(4):pp. 341–359, 1997.
7.K. Lee and Z. Geem, “A new structural optimization method based on the harmony search algorithm,” Computers and Structures, 82:781-98, 2004. 1.M. Melanie, “An Introduction to Genetic Algorithms,” Massachusett's MIT Press, 34(7):pp.1-9, 1999.
2.L. A. Ingber, “Simulated annealing: practice versus theory,” J. Math. Comput. Modell., 18(11):pp. 29–57, 1993.
3.Kennedy, J., Eberhart, R.C.: Particle Swarm Optimization. Proceedings of IEEE, 1942– 1948 (1995)
4.M. Dorigo, V. Maniezzo and A. Colorni, “The ant system: optimization by a colony of cooperating agents,” IEEE Transaction System Man Cybern, B 26(1):pp. 29–41, 1996.
5.B. Franklin and M. Bergerman, “Cultural Algorithms: Concepts and Experiments,” In Proceedings of the IEEE Congress on Evolutionary Computation, 2: pp. 1245–1251, 2000.
6.R. Storn and K. Price, “Differential evolution - a simple and efficient heuristic for global optimization over continuous spaces,” Journal of Global Optimization, 11(4):pp. 341–359, 1997.
7.K. Lee and Z. Geem, “A new structural optimization method based on the harmony search algorithm,” Computers and Structures, 82:781-98, 2004.
8.E. Rashedi, H. Nezamabadi-pour and S. Saryazdi, “A Gravitational Search Algorithm,” Information Science, Special Section on High Order Fuzzy Sets, 179(13): pp. 2232-2248, 2009.
9.Brownlee, J., Clever Algorithms: Nature-Inspired Pro Recipes, LuLu Enterprises Incorporated, (1st Edition), 2011.
10.Kaveh A, Talatahari S. Novel heuristic optimization method: Charged system search, Acta Mechanica, doi: 10.1007/s00707-009-0270-4, 2010.
11.Atashpaz-Gargari, E. and Lucas, C., "Imperialist Competitive Algorithm: An Algorithm for Optimization Inspired by Imperialistic Competition", IEEE Congress on Evolutionary Computation (CEC), pp. 4661-4667, 2007.
12.Talatahari, S. and Farahmand Azar, B. and Sheikholeslami, R. and Gandomi, A.H., "Imperialist competitive algorithm combined with chaos for global optimization", Commun. Nonl. Sci. Numer. Simulat., vol. 17, pp. 1312-1319, 2012.
13.Kaveh, A. and Talataheri, S., "Optimum design of skeletal structures using imperialist competitive algorithm", computers & structures journal, vol. 88(21), pp. 1220-1229, 2010.
14.H. Bahrami, K. Faez, M. Abdechiri, “Imperialist Competitive Algorithm using Chaos Theory for Optimization,” UKSim-AMSS 12th International Conference on Computer Modeling and Simulation, 2010.
15.M. Abdechiri, K. Faez and H. Bahrami, “Neural Network Learning based on Chaotic Imperialist Competitive Algorithm,” The 2nd International Workshop on Intelligent System and Applications (ISA2010), 2010.
16.Abdechiri, M. and Faez, K. and Bahrami, H., "Adaptive Imperialist Competitive Algorithm (AICA)", 9th IEEE International Conference on Cognitive Informatics (ICCI), pp. 940-945, 2010.
17.Coelho, L. D. and Afonso, L. and Alotto, P., "A Modified Imperialist Competitive Algorithm for Optimization in Electromagnetics", IEEE Transactions on Magnetics, vol. 48, pp. 579-582, 2012
14.H. Bahrami, K. Faez, M. Abdechiri, “Imperialist Competitive Algorithm using Chaos Theory for Optimization,” UKSim-AMSS 12th International Conference on Computer Modeling and Simulation, 2010.
15.M. Abdechiri, K. Faez and H. Bahrami, “Neural Network Learning based on Chaotic Imperialist Competitive Algorithm,” The 2nd International Workshop on Intelligent System and Applications (ISA2010), 2010.
16.Abdechiri, M. and Faez, K. and Bahrami, H., "Adaptive Imperialist Competitive Algorithm (AICA)", 9th IEEE International Conference on Cognitive Informatics (ICCI), pp. 940-945, 2010.
17.Coelho, L. D. and Afonso, L. and Alotto, P., "A Modified Imperialist Competitive Algorithm for Optimization in Electromagnetics", IEEE Transactions on Magnetics, vol. 48, pp. 579-582, 2012
14.H. Bahrami, K. Faez, M. Abdechiri, “Imperialist Competitive Algorithm using Chaos Theory for Optimization,” UKSim-AMSS 12th International Conference on Computer Modeling and Simulation, 2010.
15.M. Abdechiri, K. Faez and H. Bahrami, “Neural Network Learning based on Chaotic Imperialist Competitive Algorithm,” The 2nd International Workshop on Intelligent System and Applications (ISA2010), 2010.
16.Abdechiri, M. and Faez, K. and Bahrami, H., "Adaptive Imperialist Competitive Algorithm (AICA)", 9th IEEE International Conference on Cognitive Informatics (ICCI), pp. 940-945, 2010.
17.Coelho, L. D. and Afonso, L. and Alotto, P., "A Modified Imperialist Competitive Algorithm for Optimization in Electromagnetics", IEEE Transactions on Magnetics, vol. 48, pp. 579-582, 2012
18.Soltani-Sarvestani, M. A. and Lotfi, S. and Ramezani, F., "Quad Countries Algorithm (QCA)", Lecture Notes in Computer Science, vol. 7198, pp. 119-129, 2012.
19.Kaveh, A. and Talataheri, S., "Imperialist Competitive Algorithm For Engineering Design Problems", Asian Journal Of Civil Engineering, vol. 11, pp. 675-697, 2010.
20.Sigillito, V.G., Wing, S.P., Hutton, L.V., and Baker, K.B. (1989), ‘Classification of Radar Returns from the Ionosphere using Neural Networks’, Johns Hopkins APL Technical Digest, 10, 262–266.
21.Gorman, R.P., and Sejnowski, T.J. (1988), ‘Analysis of Hidden Units in a Layered Network Trained to Classify Sonar Targets’, Neural Networks, 1, 75–89.