The Orb-Weaving Spider Algorithm for Training of Recurrent Neural Networks

Описание

Тип публикации: статья из журнала

Год издания: 2022

Идентификатор DOI: 10.3390/sym14102036

Ключевые слова: artificial intelligence, development and information communication technology, global optimization problem, meta-european algorithms, neural network weighting adjustment, recurrent neural networks, resource-use efficiency, spider-cycle algorithm

Аннотация: The quality of operation of neural networks in solving application problems is determined by the success of the stage of their training. The task of learning neural networks is a complex optimization task. Traditional learning algorithms have a number of disadvantages, such as «sticking» in local minimums and a low convergence rateПоказать полностью. Modern approaches are based on solving the problems of adjusting the weights of neural networks using metaheuristic algorithms. Therefore, the problem of selecting the optimal set of values of algorithm parameters is important for solving application problems with symmetry properties. This paper studies the application of a new metaheuristic optimization algorithm for weights adjustment—the algorithm of the spiders-cycle, developed by the authors of this article. The approbation of the proposed approach is carried out to adjust the weights of recurrent neural networks used to solve the time series forecasting problem on the example of three different datasets. The results are compared with the results of neural networks trained by the algorithm of the reverse propagation of the error, as well as three other metaheuristic algorithms: particle swarm optimization, bats, and differential evolution. As performance criteria for the comparison of algorithms of global optimization, in this work, descriptive statistics for metrics of the estimation of quality of predictive models, as well as the number of calculations of the target function, are used. The values of the MSE and MAE metrics on the studied datasets were obtained by adjusting the weights of the neural networks using the cycling spider algorithm at 1.32, 25.48, 8.34 and 0.38, 2.18, 1.36, respectively. Compared to the inverse error propagation algorithm, the cycling spider algorithm reduced the value of the error metrics. According to the results of the study, it is concluded that the developed algorithm showed high results and, in the assessment of performance, was not inferior to the existing algorithm. © 2022 by the authors.

Ссылки на полный текст

Издание

Журнал: Symmetry

Выпуск журнала: Vol. 14, Is. 10

Номера страниц: 2036

ISSN журнала: 20738994

Издатель: MDPI

Персоны

  • Mikhalev A.S. (Scientific and Educational Software Laboratory of Informatics Department, Institute of Space and Information Technologies, Siberian Federal University, Krasnoyarsk, 660041, Russian Federation, Digital Material Science: New Materials and Technologies, Bauman Moscow State Technical University, Moscow, 105005, Russian Federation)
  • Tynchenko V.S. (Digital Material Science: New Materials and Technologies, Bauman Moscow State Technical University, Moscow, 105005, Russian Federation, Department of Technological Machines and Equipment of Oil and Gas Complex, School of Petroleum and Natural Gas Engineering, Siberian Federal University, Krasnoyarsk, 660041, Russian Federation, Information-Control Systems Department, Institute of Computer Science and Telecommunications, Reshetnev Siberian State University of Science and Technology, Krasnoyarsk, 660037, Russian Federation)
  • Nelyub V.A. (Digital Material Science: New Materials and Technologies, Bauman Moscow State Technical University, Moscow, 105005, Russian Federation)
  • Lugovaya N.M. (Scientific and Educational Software Laboratory of Informatics Department, Institute of Space and Information Technologies, Siberian Federal University, Krasnoyarsk, 660041, Russian Federation)
  • Baranov V.A. (Scientific and Educational Software Laboratory of Informatics Department, Institute of Space and Information Technologies, Siberian Federal University, Krasnoyarsk, 660041, Russian Federation)
  • Kukartsev V.V. (Digital Material Science: New Materials and Technologies, Bauman Moscow State Technical University, Moscow, 105005, Russian Federation, Department of Informatics, Institute of Space and Information Technologies, Siberian Federal University, Krasnoyarsk, 660041, Russian Federation, Department of Information Economic Systems, Institute of Engineering and Economics, Reshetnev Siberian State University of Science and Technology, Krasnoyarsk, 660037, Russian Federation)
  • Sergienko R.B. (Machine Learning Department, Gini Gmbh, Munich, 80339, Germany)
  • Kurashkin S.O. (Digital Material Science: New Materials and Technologies, Bauman Moscow State Technical University, Moscow, 105005, Russian Federation, Information-Control Systems Department, Institute of Computer Science and Telecommunications, Reshetnev Siberian State University of Science and Technology, Krasnoyarsk, 660037, Russian Federation, Laboratory of Biofuel Compositions, Siberian Federal University, Krasnoyarsk, 660041, Russian Federation)

Вхождение в базы данных