Journals / El-Cezeri / 2020 / Cilt: 7 - Sayı: 3

Effect of topological changes of resilient small-world networks on performance in the function approach problem

Fonksiyon yaklaşımı probleminde esnek küçük-dünya ağlarının topolojik değişiminin performansa etkisi

Journal
El-Cezeri
Pages
1398–1406
Publication Date
2020-09-30
DOI
—
🏛️Bibliographic metadata and full text access for this article are provided via DergiPark.
DergiPark Record ↗
✨

AI Summary & Key Takeaways

AI Insight
⚡ TL;DR(Too Long; Didn't Read)

The results show that the resilient small-world network improves the performance much more than the traditional resilient artificial neural network topology and it has been observed that increasing the number of layers significantly contributes to the network performance.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:Performance of Feed forward neural networks is directly related to topology.
📊
Key Findings & Data:In recent years, studies on improving performance of network with topological changes have become a very important process.
💡
Conclusion & Impact:The results show that the resilient small-world network improves the performance much more than the traditional resilient artificial neural network topology and it has been observed that increasing the number of layers significantly contributes to the network performance.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Newman-Watts algoritması · esnek küçük-dünya ağları · fonksiyon yaklaşımı problemi

Abstract

Performance of Feed forward neural networks is directly related to topology. In recent years, studies on improving performance of network with topological changes have become a very important process. Therefore, the small-world network model, an approach in which synaptic connections between neurons are altered, has an important place for future studies. In this study, how the performance of the resilient small-world network model changes with layered and neuronal revisions has been investigated for function approach problems. The results show that the resilient small-world network improves the performance much more than the traditional resilient artificial neural network topology and it has been observed that increasing the number of layers significantly contributes to the network performance. On the other hand, It has been observed that when the number of hidden layer neurons is increased with the number of layers, the performance has a negative effect. In this context, it has been shown that resilient small-world networks with multiple hidden layers have a positive effect on the solution of the function approach problem and also it is emphasized that this model will play a key role in improving the performance of deep neuron networks in the future.

Özet

İleri yönlü yapay sinir ağları performansı topolojiyle doğrudan ilişkilidir. Son yıllarda, topolojik değişikliklerle ağ performansının artırılması üzerine yapılan çalışmalar çok önemli bir süreç haline gelmiştir. Bu nedenle, nöronlar arası sinaptik bağlantıların değiştirildiği bir yaklaşım olan küçük-dünya ağ modeli gelecek çalışmalar için önemli bir yere sahiptir. Bu çalışmada, fonksiyon yaklaşımı problemleri için esnek küçük-dünya ağ modelinin katmansal ve nöronal değişimlerle performansının nasıl değiştiği araştırılmıştır. Elde edilen sonuçlar gösterdi ki, geleneksel esnek yapay sinir ağı topolojisine göre esnek küçük-dünya ağının performansı çok daha fazla iyileştirdiği ve katman sayısının artırılmasının ağ performansına ciddi katkı sağladığı gözlemlenmiştir. Diğer taraftan, katman sayısı ile birlikte gizli katman nöron sayısının artırıldıkça performansa negatif etki ettiği görülmüştür. Bu bağlamda, çoklu gizli katmana sahip esnek küçük-dünya ağlarının fonksiyon yaklaşımı probleminin çözümünde pozitif etki ettiği gösterilmiş ve özellikle bu modelin gelecekte derin nöron ağları performansının geliştirilmesinde anahtar rol oynayacağı vurgulanmıştır.

🔖

Cite This Article

ERKAYMAZ, O. (2020). Effect of topological changes of resilient small-world networks on performance in the function approach problem. El-Cezeri, Cilt: 7 - Sayı: 3, 1398–1406. https://ajindex.com/articles/252363

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (17)

  1. Haykin, S., Neural networks—a comprehensive foundation, (2nd Edition), Prentice-Hall, Englewood Cliffs, NJ., 1999
  2. Magnitskii N A., Some New Approaches to the Construction and Learning of Artificial neural Networks, Computational Mathematics and Modeling, 2001,12, 293-304
  3. Sun, M., Stam, A., Steuer, RE., Solving Multiple Objective Programming Problems Using Feed-Forward Artificial Neural Networks: The Interactive FFANN Procedure, Man. Sci., 1996, 42(6), 835-849
  4. Erkaymaz, O., Ozer, M & Yumusak, N., Impact of small-world topology on the performance of a feed-forward artificial neural network based on 2 different real-life problems, Turk. J. Elec. Eng. & Comp. Sci., 2014, 22, 708-718
  5. Riedmiller, M. and Braun, H., A direct adaptive method for faster backpropagation learning: The RPROP Algorithm, Proceedings of the IEEE International Conference on Neural Networks. 1993, 586-591
  6. Riedmiller, M., Braun, H., Neural speed controller trained online by means of modified rprop algorithm, IEEE Trans. Ind. Inform, 2015, 11, 586–591
  7. Shrestha, S. B. & Song, Q., Robust learning in SpikeProp. Neural Networks, 2017, 86, 54-68
  8. Yilmaz, E., Baysal, V., Ozer, M., Perc, M. Autaptic pacemaker mediated propagation of weak rhythmic activity across small-world neuronal networks, Phys. A, 2016, 444, 538-546
  9. Yilmaz, E., Baysal, V., Perc, M., Ozer, M., Enhancement of pacemaker induced stochastic resonance by an autapse in a scale-free neuronal network, Sci. China Tech. Sci., 2016, 59(3),364-370
  10. Watts, D. J. & Strogatz, S. H., Collective dynamics of 'small-world' networks, Nature, 1999, 393, 409-10
  11. Watts D. J., Small Worlds: The Dynamics of networks between order and randomness, Princeton, NJ:Princeton University Press, 2003
  12. Newman, M. E. J. & Watts, D. J., Scaling and percolation in the small-world network model, Phys. Rev. E, 1999, 60, 7332-7342
  13. Kawai, Y., Park, J. & Asada, M., A small-world topology enhances the echo state property and signal propagation in reservoir computing, Neural Networks, 2019, 112, 15-23
  14. Barabasi, AL., Albert, R., Emergence of Scaling in Random Networks, Science, 1999, 286, 509-512
  15. Simard, D., Nadeau, L. & Kröger, H., Fastest learning in small-world neural networks, Physics Letters A., 2005, 336(1),8-15
  16. Shuzhong, Y., Siwei, L., Jianyu, Li., Building Multi-layer Small World Neural Network, Lect. Note in Comp. Sci. Ser., 2006, 3971, 695-700
  17. Erkaymaz, O., Ozer, M., Impact of s