The Parallel Modification to the Levenberg-Marquardt Algorithm

被引:9
|
作者
Bilski, Jaroslaw [1 ]
Kowalczyk, Bartosz [1 ]
Grzanek, Konrad [2 ,3 ]
机构
[1] Czestochowa Tech Univ, Inst Computat Intelligence, Czestochowa, Poland
[2] Univ Social Sci, Informat Technol Inst, Lodz, Poland
[3] Clark Univ, Worcester, MA 01610 USA
关键词
Feed-forward neural network; Parallel neural network training algorithm; Optimization problem; Levenberg-Marquardt algorithm; QR decomposition; Givens rotation;
D O I
10.1007/978-3-319-91253-0_2
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The paper presents a parallel approach to the Levenberg-Marquardt algorithm (also called LM or LMA). The first section contains the mathematical basics of the classic LMA. Then the parallel modification to LMA is introduced. The classic Levenberg-Marquardt algorithm is sufficient for a training of small neural networks. For bigger networks the algorithm complexity becomes too big for the effective teaching. The main scope of this paper is to propose more complexity efficient approach to LMA by parallel computation. The proposed modification to LMA has been tested on a few function approximation problems and has been compared to the classic LMA. The paper concludes with the resolution that the parallel modification to LMA could significantly improve algorithm performance for bigger networks. Summary also contains a several proposals for the possible future work directions in the considered area.
引用
收藏
页码:15 / 24
页数:10
相关论文
共 50 条
  • [1] A Parallel Levenberg-Marquardt Algorithm
    Cao, Jun
    Novstrup, Krista A.
    Goyal, Ayush
    Midkiff, Samuel R.
    Caruthers, James M.
    [J]. ICS'09: PROCEEDINGS OF THE 2009 ACM SIGARCH INTERNATIONAL CONFERENCE ON SUPERCOMPUTING, 2009, : 450 - 459
  • [2] Parallel and separable recursive Levenberg-Marquardt training algorithm
    Asirvadam, VS
    McLoone, SF
    Irwin, GW
    [J]. NEURAL NETWORKS FOR SIGNAL PROCESSING XII, PROCEEDINGS, 2002, : 129 - 138
  • [3] Parallel Levenberg-Marquardt Algorithm Without Error Backpropagation
    Bilski, Jaroslaw
    Wilamowski, Bogdan M.
    [J]. ARTIFICIAL INTELLIGENCE AND SOFT COMPUTING, ICAISC 2017, PT I, 2017, 10245 : 25 - 39
  • [4] Adaptive Levenberg-Marquardt Algorithm: A New Optimization Strategy for Levenberg-Marquardt Neural Networks
    Yan, Zhiqi
    Zhong, Shisheng
    Lin, Lin
    Cui, Zhiquan
    [J]. MATHEMATICS, 2021, 9 (17)
  • [5] Parallel Approach to the Levenberg-Marquardt Learning Algorithm for Feedforward Neural Networks
    Bilski, Jaroslaw
    Smolag, Jacek
    Zurada, Jacek M.
    [J]. ARTIFICIAL INTELLIGENCE AND SOFT COMPUTING, PT I, 2015, 9119 : 3 - 14
  • [6] The application and modeling of the Levenberg-Marquardt algorithm
    Li, Jian-rong
    [J]. 2010 2ND INTERNATIONAL CONFERENCE ON E-BUSINESS AND INFORMATION SYSTEM SECURITY (EBISS 2010), 2010, : 278 - 280
  • [7] Distributed localization using Levenberg-Marquardt algorithm
    Shervin Parvini Ahmadi
    Anders Hansson
    Sina Khoshfetrat Pakazad
    [J]. EURASIP Journal on Advances in Signal Processing, 2021
  • [8] A Novel Modification on the Levenberg-Marquardt Algorithm for Avoiding Overfitting in Neural Network Training
    Iplikci, Serdar
    Bilgi, Batuhan
    Menemen, Ali
    Bahtiyar, Bedri
    [J]. ARTIFICIAL NEURAL NETWORKS AND MACHINE LEARNING - ICANN 2019: DEEP LEARNING, PT II, 2019, 11728 : 201 - 207
  • [9] A Levenberg-Marquardt algorithm for unconstrained multicriteria optimization
    Fischer, Andreas
    Shukla, Pradyumn K.
    [J]. OPERATIONS RESEARCH LETTERS, 2008, 36 (05) : 643 - 646
  • [10] Convergence analysis of a subsampled Levenberg-Marquardt algorithm
    Xing, Ganchen
    Gu, Jian
    Xiao, Xiantao
    [J]. OPERATIONS RESEARCH LETTERS, 2023, 51 (04) : 379 - 384