Local search and pseudoinversion: an hybrid approach to neural network training

被引：1
|
作者：
Rubini, Luca [1 ]
Cancelliere, Rossella [1 ]
Gallinari, Patrick [2 ]
Grosso, Andrea [1 ]
机构：
[1] Univ Turin, Dept Comp Sci, Turin, Italy
[2] Univ Paris 06, Comp Sci Lab, LIP6, Paris, France
来源：
KNOWLEDGE AND INFORMATION SYSTEMS | 2016年 / 48卷 / 02期
关键词：
Neural networks; Random projections; Local search; Pseudoinverse matrix; EXTREME LEARNING-MACHINE; SPECIAL-ISSUE;
D O I：
10.1007/s10115-016-0935-y
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
We consider recent successful techniques proposed for neural network training that set randomly the weights from input to hidden layer, while weights from hidden to output layer are analytically determined by Moore-Penrose generalized inverse. This study aimed to analyse the impact on performances when the completely random sampling of the space of input weights is replaced by a local search procedure over a discretized set of weights. The performances of the proposed training methods are assessed through computational experience on several UCI datasets.
引用
收藏
页码：493 / 503
页数：11
相关论文
共 50 条
  • [1] Local search and pseudoinversion: an hybrid approach to neural network training
    Luca Rubini
    Rossella Cancelliere
    Patrick Gallinari
    Andrea Grosso
    Knowledge and Information Systems, 2016, 48 : 493 - 503
  • [2] Optimization of the local search in the training for SAMANN neural network
    Medvedev, Viktor
    Dzemyda, Gintautas
    JOURNAL OF GLOBAL OPTIMIZATION, 2006, 35 (04) : 607 - 623
  • [3] Optimization of the Local Search in the Training for SAMANN Neural Network
    Viktor Medvedev
    Gintautas Dzemyda
    Journal of Global Optimization, 2006, 35 : 607 - 623
  • [4] Job-shop scheduling with an adaptive neural network and local search hybrid approach
    Yang, Shengxiang
    2006 IEEE INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORK PROCEEDINGS, VOLS 1-10, 2006, : 2720 - 2727
  • [5] Design of manufacturing systems by a hybrid approach with neural network metamodelling and stochastic local search
    Chen, MC
    Yang, TH
    INTERNATIONAL JOURNAL OF PRODUCTION RESEARCH, 2002, 40 (01) : 71 - 92
  • [6] Hybrid neural networks: An evolutionary approach with local search
    Iyoda, EM
    Von Zuben, FJ
    INTEGRATED COMPUTER-AIDED ENGINEERING, 2002, 9 (01) : 57 - 72
  • [7] An analysis of numerical issues in neural training by pseudoinversion
    Cancelliere, R.
    Deluca, R.
    Gai, M.
    Gallinari, P.
    Rubini, L.
    COMPUTATIONAL & APPLIED MATHEMATICS, 2017, 36 (01): : 599 - 609
  • [8] An analysis of numerical issues in neural training by pseudoinversion
    R. Cancelliere
    R. Deluca
    M. Gai
    P. Gallinari
    L. Rubini
    Computational and Applied Mathematics, 2017, 36 : 599 - 609
  • [9] A Hybrid of Cuckoo Search and Simplex Method for Fuzzy Neural Network Training
    Chang, Jyh-Yeong
    Liao, Shih-Hui
    Wu, Shang-Lin
    Lin, Chin-Teng
    2015 IEEE 12TH INTERNATIONAL CONFERENCE ON NETWORKING, SENSING AND CONTROL (ICNSC), 2015, : 13 - 16
  • [10] A hybrid Gravitational Search Algorithm-Genetic Algorithm for neural network training
    Sheikhpour, Saeide
    Sabouri, Mahdieh
    Zahiri, Seyed-Hamid
    2013 21ST IRANIAN CONFERENCE ON ELECTRICAL ENGINEERING (ICEE), 2013,
12345