SavedAdd your journal

Journals / EMO Bilimsel Dergi / 2021 / Cilt: 11 - Sayı: 21

A Novel Improved Atom Search Optimization Algorithm for Training Multilayer Perceptron

Geliştirilmiş Atom Arama Optimizasyon Algoritması ile Çok Katmanlı Algılayıcı Eğitimi

Davut İzci
Pages
71–79
Publication Date
2022-10-12
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 clearly indicated the performance of the proposed algorithm to be better as the lower average and standard deviation of mean square error were achieved via the proposed approach.

📌 3 Key Takeaways

🎯
Objective & Hypothesis:A novel hybrid algorithm developed by merging atom search optimization (ASO) and simulated annealing (SA) algorithms is presented.
📊
Key Findings & Data:The search capability of ASO was improved by using simulated annealing (SA) algorithm.
💡
Conclusion & Impact:The results clearly indicated the performance of the proposed algorithm to be better as the lower average and standard deviation of mean square error were achieved via the proposed approach.
🔬Design:Modeling & Simulation
📍Scope:Comprehensive Sample
⚙️Variables:Çok katmanlı algılayıcı · sezgisel-üstü algoritmalar · atom arama optimizasyonu

Abstract

A novel hybrid algorithm developed by merging atom search optimization (ASO) and simulated annealing (SA) algorithms is presented. The search capability of ASO was improved by using simulated annealing (SA) algorithm. The proposed hybrid algorithm was named as iASO and and used for training multilayer perceptron (MLP) to observe its ability for optimizing non-linear systems. Several datasets (Iris, Balloon, XOR, Breast cancer and Heart) were used, and the obtained results were compared with respective recent competitive algorithms such as original ASO, sine cosine algorithm (SCA), particle swarm optimization (PSO), ant colony optimization (ACO) and grey wolf optimization (GWO). The results clearly indicated the performance of the proposed algorithm to be better as the lower average and standard deviation of mean square error were achieved via the proposed approach.

Özet

Bu makalede atom arama optimizasyonu (ASO) ve benzetilmiş tavlama (SA) algoritmalarının hibritleştirilmesiyle geliştirilen ve iASO olarak isimlendirilen yeni bir hibrit algoritma ele alınmaktadır. SA tekniğinin kullanımı ile ASO algoritmasının arama yeteneği güçlendirilmiştir. Önerilen hibrit algoritmanın doğrusal olmayan sistemleri optimize etmedeki yeteneğini gözlemlemek üzere çok katmanlı algılayıcıyı (MLP) eğiticisi olarak kullanılmıştır. Iris, Balloon, XOR, Breast Cancer ve Heart olmak üzere çeşitli veri kümeleri kullanılmış ve elde edilen sonuçlar orijinal ASO, sinüs kosinüs algoritması (SCA), parçacık sürüsü optimizasyonu (PSO), karınca kolonisi optimizasyonu (ACO) ve gri kurt optimizasyonu (GWO) gibi rekabetçi algoritmalar kullanılarak oluşturulmuş diğer MLP eğiticileri ile karşılaştırılmıştır. Sonuçlar, önerilen yaklaşımla daha düşük ortalama kare hatasının (MSE) ortalama ve standart sapmasının elde edildiğini göstermiş ve dolayısıyla daha iyi performansının olduğunu açıkça göstermiştir.

🔖

Cite This Article

İzci, D. (2022). A Novel Improved Atom Search Optimization Algorithm for Training Multilayer Perceptron. EMO Bilimsel Dergi, Cilt: 11 - Sayı: 21, 71–79. https://ajindex.com/articles/497023

📄 Full Text PDF Viewer

📄

PDF görüntüleyici hazırlanıyor...

References (3)

  1. P. Bansal, S. Kumar, S. Pasrija, and S. Singh, “A hybrid grasshopper and new cat swarm optimization algorithm for feature selection and optimization of multi-layer perceptron,” Soft Comput., pp. 1–27, 2020, doi: 10.1007/s00500-020-04877-w. S. Haykin, Neural networks: A comprehensive foundation, 2nd editio. Prentice Hall PTR, 1999. S. Gupta and K. Deep, “A novel hybrid sine cosine algorithm for global optimization and its application to train multilayer perceptrons,” Appl. Intell., vol. 50, no. 4, pp. 993–1026, 2020, doi: 10.1007/s10489-019-01570-w. A. A. Suratgar, M. B. Tavakoli, and A. Hoseinabadi, “Modified Levenberg-Marquardt method for neural networks training,” World Acad Sci Eng Technol, vol. 6, no. 1, pp. 46–48, 2005.
  2. E. Eker, M. Kayri, S. Ekinci, and D. Izci, “Training Multi-Layer Perceptron Using Harris Hawks Optimization,” in 2020 International Congress on Human-Computer Interaction, Optimization and Robotic Applications (HORA), 2020, pp. 1–5, doi: 10.1109/HORA49412.2020.9152874. H. Faris, S. Mirjalili, and I. Aljarah, “Automatic selection of hidden neurons and weights in neural networks using grey wolf optimizer based on a hybrid encoding scheme,” Int. J. Mach. Learn. Cybern., vol. 10, no. 10, pp. 2901–2920, 2019, doi: 10.1007/s13042-018-00913-2.
  3. A. A. Heidari, H. Faris, S. Mirjalili, I. Aljarah, and M. Mafarja, “Ant lion optimizer: Theory, literature review, and application in multi-layer perceptron neural networks,” in Studies in Computational Intelligence, vol. 811, S. Mirjalili, J. Song Dong, and A. Lewis, Eds. Cham: Springer International Publishing, 2020, pp. 23–46. M. Khishe and M. R. Mosavi, “Classification of underwater acoustical dataset using neural network trained by Chimp Optimization Algorithm,” Appl. Acoust., vol. 157, p. 107005, 2020, doi: 10.1016/j.apacoust.2019.107005. A. A. Heidari, H. Faris, I. Aljarah, and S. Mirjalili, “An efficient hybrid multilayer perceptron neural network with grasshopper optimization,” Soft Comput., vol. 23, no. 17, pp. 7941–7958, 2019, doi: 10.1007/s00500-018-3424-2. M. Khishe and H. Mohammadi, “Passive sonar target classification using multi-layer perceptron trained by salp swarm algorithm,” Ocean Eng., vol. 181, pp. 98–108, 2019, doi: 10.1016/j.oceaneng.2019.04.013. D. Bairathi and D. Gopalani, “Numerical optimization and feed-forward neural networks training using an improved optimization algorithm: multiple leader salp swarm algorithm,” Evol. Intell., pp. 1–17, 2019, doi: 10.1007/s12065-019-00269-