Differential Evolution-based Neural Network Training Incorporating a Centroid-based Strategy and Dynamic Opposition-based Learning

被引:11
作者
Mousavirad, Seyed Jalaleddin [1 ]
Oliva, Diego [2 ]
Hinojosa, Salvador [2 ]
Schaefer, Gerald [3 ]
机构
[1] Hakim Sabzevari Univ, Dept Comp Engn, Sabzevar, Iran
[2] Univ Guadalajara, Dept Ciencias Comp, Guadalajara, Jalisco, Mexico
[3] Loughborough Univ, Dept Comp Sci, Loughborough, Leics, England
来源
2021 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC 2021) | 2021年
关键词
neural network training; optimisation; differential evolution; center-based strategy; dynamic opposition-based learning; ALGORITHM; OPTIMIZATION;
D O I
10.1109/CEC45853.2021.9504801
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Training multi-layer neural networks (MLNNs), a challenging task, involves finding appropriate weights and biases. MLNN training is important since the performance of MLNNs is mainly dependent on these network parameters. However, conventional algorithms such as gradient-based methods, while extensively used for MLNN training, suffer from drawbacks such as a tendency to getting stuck in local optima. Population-based metaheuristic algorithms can be used to overcome these problems. In this paper, we propose a novel MLNN training algorithm, CenDE-DOBL, that is based on differential evolution (DE), a centroid-based strategy (Cen-S), and dynamic opposition-based learning (DOBL). The Cen-S approach employs the centroid of the best individuals as a member of population, while other members are updated using standard crossover and mutation operators. This improves exploitation since the new member is obtained based on the best individuals, while the employed DOBL strategy, which uses the opposite of an individual, leads to enhanced exploration. Our extensive experiments compare CenDE-DOBL to 26 conventional and population-based algorithms and confirm it to provide excellent MLNN training performance.
引用
收藏
页码:1233 / 1240
页数:8
相关论文
共 50 条
  • [21] An Opposition-Based Hybrid Artificial Bee Colony with Differential Evolution
    Worasucheep, Chukiat
    [J]. 2015 IEEE CONGRESS ON EVOLUTIONARY COMPUTATION (CEC), 2015, : 2611 - 2618
  • [22] Bayesian network structure learning by opposition-based learning
    Baodan Sun
    Xinyi Zhang
    Junhui Jiang
    Jianguang Gong
    Dan Lin
    [J]. Scientific Reports, 15 (1)
  • [23] A dual opposition-based learning for differential evolution with protective mechanism for engineering optimization problems
    Li, Jiahang
    Gao, Yuelin
    Wang, Kaiguang
    Sun, Ying
    [J]. APPLIED SOFT COMPUTING, 2021, 113 (113)
  • [24] An Improvement of Opposition-Based Differential Evolution with Archive Solutions
    Kushida, Jun-ichi
    Hara, Akira
    Takahama, Tetsuyuki
    [J]. 2014 INTERNATIONAL CONFERENCE ON ADVANCED MECHATRONIC SYSTEMS (ICAMECHS), 2014, : 463 - 468
  • [25] Neighborhood opposition-based differential evolution with Gaussian perturbation
    Zhao, Xinchao
    Feng, Shuai
    Hao, Junling
    Zuo, Xingquan
    Zhang, Yong
    [J]. SOFT COMPUTING, 2021, 25 (01) : 27 - 46
  • [26] Opposition-Based Differential Evolution with Protective Generation Jumping
    Esmailzadeh, Ali
    Rahnamayan, Shahryar
    [J]. 2011 IEEE SYMPOSIUM ON DIFFERENTIAL EVOLUTION (SDE), 2011, : 57 - 64
  • [27] Opposition-based differential evolution for hydrothermal power system
    Jagat Kishore Pattanaik
    Mousumi Basu
    Deba Prasad Dash
    [J]. Protection and Control of Modern Power Systems, 2017, 2 (1)
  • [28] Generalised opposition-based differential evolution: an experimental study
    Wang, Hui
    Rahnamayan, Shahryar
    Zeng, Sanyou
    [J]. INTERNATIONAL JOURNAL OF COMPUTER APPLICATIONS IN TECHNOLOGY, 2012, 43 (04) : 311 - 319
  • [29] Opposition-based Ensemble Micro-Differential Evolution
    Salehinejad, Hojjat
    Rahnamayan, Shahryar
    Tizhoosh, Hamid R.
    [J]. 2017 IEEE SYMPOSIUM SERIES ON COMPUTATIONAL INTELLIGENCE (SSCI), 2017, : 1128 - 1135
  • [30] Enhanced differential evolution with generalised opposition-based learning and orientation neighbourhood mining
    Wang, Jing
    [J]. INTERNATIONAL JOURNAL OF COMPUTING SCIENCE AND MATHEMATICS, 2015, 6 (01) : 49 - 58