Konferans Bildirisi
BibTex RIS Kaynak Göster

Göç Operatörü ile Yenilenen Uyarlanabilir Genetik Algoritma

Yıl 2021, Sayı: 26 - Ejosat Özel Sayı 2021 (HORA), 383 - 388, 31.07.2021
https://doi.org/10.31590/ejosat.959683

Öz

Bu çalışmada, karmaşık sistemlerin modellenmesinde karşılaşılan optimizasyon problemlerine yönelik çözüm üretmek amacıyla Genetik Algoritma ve geliştirilen Uyarlanabilir Genetik Algoritma kullanılmıştır. Bu algoritmaları kullanırken karşılaşılan en önemli problem, bu algoritmaların yerel en iyilere takılmasıdır. Örnegin, Genetik Algoritma kullanıldı ğında karmaşık sistem veri optimizasyon işleminde, belirli bir süre içinde yerel en iyilere sıklıkla takıldıgı gözlemlenmiştir. Bunun nedeni, sınırlı sayıdaki iterasyonlarda çok fazla mutasyon gerçekleştirilemeyip birey sayısının sınırlı olmasından dolayı kısa sürede popülasyonun aynı çözüm kümesi ile dolmasıdır. Buradan yola çıkarak Genetik ve Uyarlanabilir Genetik Algoritmalara göç operatörü ekleyerek daha geniş alanda arama yapmaları ve yerel en iyilerden kaçınmaları saglanmıştır. Bu çalışmada Lotka Volterra modeli kullanılarak göç operatörünün kullanıldıgı bu algoritmalar test edilmiştir. Elde edilen sonuçlar incelendiginde göç operatörünün eklendigi Uyarlanabilir Genetik Algoritmanın Genetik Algoritmadan çok daha başarılı oldugu gözlemlenmiş olup, karmaşık sistem optimizasyonunda hedeflenen başarı yakalanmıştır. Çalışmanın devamında Metot bölümünde kullanılan algoritmalar ana hatları ile açıklanmaktadır. Deneysel sonuçlar bölümünde kullanılan farklı algoritmalar lotka-voltera modelinde ve nümerik test fonksiyonların test edilip birbirleri ile karşılaştırılmaktadır. Sonuç ve tartışma bölümünde genel sonuçlar ve gelecekte yapılabilecek çalışmalar hakkında kısaca bilgi verilmektedir.

Kaynakça

  • Birattari, M., Yuan, Z. Balaprakash, P., & Stützle, T. (2010). Experimental Methods for the Analysis of Optimization Algorithms, Springer-Verlag Berlin Heidelberg.
  • Calvez B., & Hutzler G. 2007. Adaptive Dichotomic Optimization: a New Method for the Calibration Of Agent Based Models, 21st Annual European Simulation and Modelling Conference (ESM 2007), Malta, 415-419.
  • Di Marzo Serugendo, G., Gleizes, M.P., & Karageorgos, A. (2011). Self-organising software from natural to artificial adaptation, first editör, Heidelberg, Berlin: Springer-Verlag.
  • Fığlalı A., & Engin O. (2002). Genetik algoritmalarla akış tipi çizelgelemede üreme yöntemi optimizasyonu, İTÜ Dergisi, 1-6.
  • Guivarch, V., Camps, V., & Peninou, A. (2012). AMADEUS: an adaptive multi-agent system to learn a user’s recurring actions in ambient systems, Advances in Distributed Computing and Artificial Intelligence Journal, 3 (1): 1-10.
  • Imbault, F., & Lebart, K. (2004). A stochastic optimization approach for parameter tuning of support vector machines, Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR, Cambridge, 597- 600.
  • Jang, J.S.R. (1997). Neuro-Fuzzy and soft computing, A Computational Approach To Learning and Machine Intelligence, Derivative- Free Optimization, Prentice-Hall, USA.
  • Kaddoum, E. & George, J.P. (2012). Collective self-tuning for complex product design (short paper), in IEEE International Conference on Self- Adaptive and Self-Organizing Systems(SASO), Lyon, CPS, (electronic medium).
  • Korkmaz Tan, R., & Bora, ¸ S. (2017a). Parameter tuning algorithms in modeling and simulation, International Journal Of Engineering Science and Application, l (2): 58-66.
  • Korkmaz Tan, R. & Bora, ¸ S. (2017b). Parameter tuning of complex systems modeled in agent based modeling and simulation, World Academy of Science, Engineering and Technology International Journal of Computer and Information Engineering, 11 (12): 1301-1310.
  • Korkmaz Tan, R., & Bora, Ş. (2019). Adaptive parameter tuning for agent-based modeling and simulation. Simulation: Transactions of the Society for Modeling and Simulation International 2019; 95 (9): 771–796.
  • Korkmaz Tan, R., & Bora Ş.,(2020). Adaptive modified artificial bee colony algorithms (AMABC) for optimization of complex systems, Turkish Journal of Electrical Engineering & Computer Sciences, 28 (5): 2602-2629
  • Rebaudengo, M., & Reorda, M.S. (1992). An experimental analysis of effects of migration in parallel genetic algorithms, EWPDP93:IEEE/Euromicro Workshop on Parallel and Distributed Processing, Gran Canaria (E), Gennaio, 232-238.
  • Salwala, C. Kotrajaras, V. & Horkaew, P. (2010). Improving performance for emergent environments parameter tuning and simulation in games using GPU, Computer Science and Information Technology (ICCSIT), 2010 3rd IEEE International Conference on, 2: 37-41.

Adaptive Genetic Algorithm Renewed by Migration Operator

Yıl 2021, Sayı: 26 - Ejosat Özel Sayı 2021 (HORA), 383 - 388, 31.07.2021
https://doi.org/10.31590/ejosat.959683

Öz

In the present study, the Genetic Algorithm and the developed Adaptive Genetic Algorithm are used to solve optimization problems faced in modeling of complex systems. While using the Genetic Algorithm and the Adaptive Genetic Algorithm, the most important problem encountered is that these algorithms are prone to getting stuck in local bests. For example, in the complex system data optimization process using the Genetic Algorithm, it has been observed that it is frequently fitted to local bests in a certain period of time. The reasons are that, many mutations cannot be performed in the limited number of iterations, and because the number of individuals is limited, the population is filled with the same set of solutions in a short time. Therefore, the migration operator has been added to the Genetic Algorithm and the Adaptive Genetic algorithm in order to avoid local bests and to provide that these algorithms search in a wider area of the large search space of complex systems. In the present study, we have tested these algorithms using the migration operator in the Lotka Volterra Model. When the results are examined, it is observed that the Adaptive Genetic Algorithm with migration operator outperforms the Genetic Algorithm and the targeted success is achieved in complex system optimization. In the rest of the paper, the algorithms used in the Method Section are explained with outlines. In the Experimental Studies Section, the different algorithms are tested and compared with each other in the Lotka-Voltera model and numerical test functions. In the Conclusions and Discussions Section, brief information is given about general results and future studies.

Kaynakça

  • Birattari, M., Yuan, Z. Balaprakash, P., & Stützle, T. (2010). Experimental Methods for the Analysis of Optimization Algorithms, Springer-Verlag Berlin Heidelberg.
  • Calvez B., & Hutzler G. 2007. Adaptive Dichotomic Optimization: a New Method for the Calibration Of Agent Based Models, 21st Annual European Simulation and Modelling Conference (ESM 2007), Malta, 415-419.
  • Di Marzo Serugendo, G., Gleizes, M.P., & Karageorgos, A. (2011). Self-organising software from natural to artificial adaptation, first editör, Heidelberg, Berlin: Springer-Verlag.
  • Fığlalı A., & Engin O. (2002). Genetik algoritmalarla akış tipi çizelgelemede üreme yöntemi optimizasyonu, İTÜ Dergisi, 1-6.
  • Guivarch, V., Camps, V., & Peninou, A. (2012). AMADEUS: an adaptive multi-agent system to learn a user’s recurring actions in ambient systems, Advances in Distributed Computing and Artificial Intelligence Journal, 3 (1): 1-10.
  • Imbault, F., & Lebart, K. (2004). A stochastic optimization approach for parameter tuning of support vector machines, Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR, Cambridge, 597- 600.
  • Jang, J.S.R. (1997). Neuro-Fuzzy and soft computing, A Computational Approach To Learning and Machine Intelligence, Derivative- Free Optimization, Prentice-Hall, USA.
  • Kaddoum, E. & George, J.P. (2012). Collective self-tuning for complex product design (short paper), in IEEE International Conference on Self- Adaptive and Self-Organizing Systems(SASO), Lyon, CPS, (electronic medium).
  • Korkmaz Tan, R., & Bora, ¸ S. (2017a). Parameter tuning algorithms in modeling and simulation, International Journal Of Engineering Science and Application, l (2): 58-66.
  • Korkmaz Tan, R. & Bora, ¸ S. (2017b). Parameter tuning of complex systems modeled in agent based modeling and simulation, World Academy of Science, Engineering and Technology International Journal of Computer and Information Engineering, 11 (12): 1301-1310.
  • Korkmaz Tan, R., & Bora, Ş. (2019). Adaptive parameter tuning for agent-based modeling and simulation. Simulation: Transactions of the Society for Modeling and Simulation International 2019; 95 (9): 771–796.
  • Korkmaz Tan, R., & Bora Ş.,(2020). Adaptive modified artificial bee colony algorithms (AMABC) for optimization of complex systems, Turkish Journal of Electrical Engineering & Computer Sciences, 28 (5): 2602-2629
  • Rebaudengo, M., & Reorda, M.S. (1992). An experimental analysis of effects of migration in parallel genetic algorithms, EWPDP93:IEEE/Euromicro Workshop on Parallel and Distributed Processing, Gran Canaria (E), Gennaio, 232-238.
  • Salwala, C. Kotrajaras, V. & Horkaew, P. (2010). Improving performance for emergent environments parameter tuning and simulation in games using GPU, Computer Science and Information Technology (ICCSIT), 2010 3rd IEEE International Conference on, 2: 37-41.
Toplam 14 adet kaynakça vardır.

Ayrıntılar

Birincil Dil İngilizce
Konular Mühendislik
Bölüm Makaleler
Yazarlar

Rabia Korkmaz Tan 0000-0002-3777-2536

Şebnem Bora 0000-0003-0111-4635

Yayımlanma Tarihi 31 Temmuz 2021
Yayımlandığı Sayı Yıl 2021 Sayı: 26 - Ejosat Özel Sayı 2021 (HORA)

Kaynak Göster

APA Korkmaz Tan, R., & Bora, Ş. (2021). Adaptive Genetic Algorithm Renewed by Migration Operator. Avrupa Bilim Ve Teknoloji Dergisi(26), 383-388. https://doi.org/10.31590/ejosat.959683