Training multilayer perceptrons via interpolation by superpositions of an activation function

被引：0
|
作者：
Li, X [1 ]
机构：
[1] Univ Nevada, Dept Math Sci, Las Vegas, NV 89154 USA
来源：
METMBS'01: PROCEEDINGS OF THE INTERNATIONAL CONFERENCE ON MATHEMATICS AND ENGINEERING TECHNIQUES IN MEDICINE AND BIOLOGICAL SCIENCES | 2001年
关键词：
multilayer perceptrons; network training; interpolation;
D O I：
暂无
中图分类号：
Q5 [生物化学];
学科分类号：
071010 ; 081704 ;
摘要：
It is known (cf. [9]) that the problem of training feedforward multilayer perceptrons is equivalent to the problem of interpolating the given sample data by the superpositions of activation functions of the networks. In this paper we explicitly construct the invertible matrices associated with interpolation by using sigmoids as the activation functions, and also construct the invertible matrices for smooth non-sigmoid functions under certain mild conditions.
引用
收藏
页码：19 / 24
页数：6
相关论文
共 50 条
  • [1] AN ALGORITHM FOR TRAINING MULTILAYER PERCEPTRONS FOR DATA CLASSIFICATION AND FUNCTION INTERPOLATION
    MADYASTHA, RK
    AAZHANG, B
    IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS I-FUNDAMENTAL THEORY AND APPLICATIONS, 1994, 41 (12): : 866 - 875
  • [2] Training multilayer perceptrons via minimization of sum of ridge functions
    Wu, W
    Feng, GR
    Li, X
    ADVANCES IN COMPUTATIONAL MATHEMATICS, 2002, 17 (04) : 331 - 347
  • [3] Training Multilayer Perceptrons Via Minimization of Sum of Ridge Functions
    Wei Wu
    Guorui Feng
    Xin Li
    Advances in Computational Mathematics, 2002, 17 : 331 - 347
  • [4] Fast training of multilayer perceptrons
    Verma, B
    IEEE TRANSACTIONS ON NEURAL NETWORKS, 1997, 8 (06): : 1314 - 1320
  • [5] A new error function at hidden layers for fast training of multilayer perceptrons
    Oh, SH
    Lee, SY
    IEEE TRANSACTIONS ON NEURAL NETWORKS, 1999, 10 (04): : 960 - 964
  • [6] An adaptive method of training multilayer perceptrons
    Lo, JT
    Bassu, D
    IJCNN'01: INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS, VOLS 1-4, PROCEEDINGS, 2001, : 2013 - 2018
  • [7] Robust formulations for training multilayer perceptrons
    Kärkkäinen, T
    Heikkola, E
    NEURAL COMPUTATION, 2004, 16 (04) : 837 - 862
  • [8] Efficient pruning of multilayer perceptrons using a fuzzy sigmoid activation function
    Soria-Olivas, E
    Martín-Guerrero, JD
    Serrano-López, AJ
    Calpe-Maravilla, J
    Vila-Francés, J
    Camps-Valls, G
    NEUROCOMPUTING, 2006, 69 (7-9) : 909 - 912
  • [9] Training multilayer perceptrons parameter by parameter
    Li, YL
    Wang, KQ
    PROCEEDINGS OF THE 2004 INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND CYBERNETICS, VOLS 1-7, 2004, : 3397 - 3401
  • [10] Interpreting Multilayer Perceptrons Using 3-Valued Activation Function
    Furukawa, Tomoya
    Zhao, Qiangfu
    2017 3RD IEEE INTERNATIONAL CONFERENCE ON CYBERNETICS (CYBCONF), 2017, : 326 - 331
12345