Convergence analysis of batch gradient algorithm for three classes of sigma-pi neural networks

被引：17

作者：

Zhang, Chao ^{[1
]}

Wu, Wei ^{[1
]}

Xiong, Yan ^{[1
]}

机构：

[1] Dalian Univ Technol, Dept Appl Math, Dalian 116024, Peoples R China

来源：

NEURAL PROCESSING LETTERS | 2007年 / 26卷 / 03期

基金：

中国国家自然科学基金;

关键词：

convergence; sigma-Pi-Sigma neural networks; sigma-Sigma-Pi neural networks; sigma-Pi-Sigma-Pi neural networks; batch gradient algorithm; monotonicity;

D O I：

10.1007/s11063-007-9050-0

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Sigma-Pi (Sigma u-Sigma a) neural networks (SPNNs) are known to provide more powerful mapping capability than traditional feed-forward neural networks. A unified convergence analysis for the batch gradient algorithm for SPNN learning is presented, covering three classes of SPNNs: Sigma u-Sigma a-Sigma u, Sigma u-Sigma u-Sigma a and Sigma u-Sigma a-Sigma u-Sigma a. The monotonicity of the error function in the iteration is also guaranteed.

引用

页码：177 / 189

页数：13

共 13 条

[1] Bertsekas D. P., 1996, Neuro Dynamic Programming, V1st
[2] Product Units: A Computationally Powerful and Biologically Plausible Extension to Backpropagation Networks
Durbin, Richard
Rumelhart, David E.
[J]. NEURAL COMPUTATION, 1989, 1 (01) : 133 - 142
[3] A FRAMEWORK FOR-IMPROVED TRAINING OF SIGMA-PI NETWORKS
HEYWOOD, M
NOAKES, P
[J]. IEEE TRANSACTIONS ON NEURAL NETWORKS, 1995, 6 (04): : 893 - 903
[4] ANALYSIS OF ADAPTIVE STEP-SIZE SA ALGORITHMS FOR PARAMETER TRACKING
KUSHNER, HJ
YANG, J
[J]. IEEE TRANSACTIONS ON AUTOMATIC CONTROL, 1995, 40 (08) : 1403 - 1410
[5] Note on interpolation on the hypercube by means of sigma-pi neural networks
Lenze, B
[J]. NEUROCOMPUTING, 2004, 61 : 471 - 478
[6] HOW TO MAKE SIGMA-PI NEURAL NETWORKS PERFORM PERFECTLY ON REGULAR TRAINING SETS
LENZE, B
[J]. NEURAL NETWORKS, 1994, 7 (08) : 1285 - 1293
[7] Successive approximation training algorithm for feedforward neural networks
Liang, YC
Feng, DP
Lee, HP
Lim, SP
Lee, KH
[J]. NEUROCOMPUTING, 2002, 42 : 311 - 322
[8] Rumelhart DE, 1986, PARALLEL DISTRIBUTED, V1, DOI DOI 10.7551/MITPRESS/5236.001.0001
[9] SHAO HM, 2004, P INT S COMP INF, V2, P627
[10] Wu W, 2005, PROCEEDINGS OF THE 2005 INTERNATIONAL CONFERENCE ON NEURAL NETWORKS AND BRAIN, VOLS 1-3, P332

← 1 2 →