A new strategy for adaptively constructing multilayer feedforward neural networks

被引：55

作者：

Ma, L ^{[1
]}

Khorasani, K ^{[1
]}

机构：

[1] Concordia Univ, Dept Elect & Comp Engn, Montreal, PQ H3G 1M8, Canada

来源：

NEUROCOMPUTING | 2003年 / 51卷

基金：

加拿大自然科学与工程研究理事会;

关键词：

constructive networks; multilayer feedforward neural networks; regression and function approximation;

D O I：

10.1016/S0925-2312(02)00597-0

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

In this paper a new strategy for adaptively and autonomously constructing a multi-hidden-layer feedforward neural network (FNN) is introduced. The proposed scheme belongs to a class of structure level adaptation algorithms that adds both new hidden units and new hidden layers one at a time when it is determined to be needed. Using this strategy, a FNN may be constructed having as many hidden layers and hidden units as required by the complexity of the problem being considered. Simulation results applied to regression problems are included to demonstrate the performance capabilities of the proposed scheme. (C) 2002 Elsevier Science B.V. All rights reserved.

引用

页码：361 / 385

页数：25

共 35 条

[1] Ash T., 1989, Connection Science, V1, P365, DOI 10.1080/09540098908915647
[2] Bose N.K., 1996, Neural Network Fundamentals with Graphs, Algorithms, and Applications
[3] Buntine W. L., 1991, Complex Systems, V5, P603
[4] ADJUVANT SYSTEMIC THERAPY FOR BREAST-CANCER IN THE ELDERLY - COMPETING CAUSES OF MORTALITY
CASTIGLIONE, M
GELBER, RD
GOLDHIRSCH, A
[J]. JOURNAL OF CLINICAL ONCOLOGY, 1990, 8 (03) : 519 - 526
[5] Castillo PA, 1999, LECT NOTES COMPUT SC, V1606, P661
[6] CHAUVIN Y, 1990, ADV NEURAL INFORMATI, V2, P642
[7] DRAELOS T, 1996, IEEE INT C NEUR NETW, V1, P50
[8] FAHLMAN S, 1991, CMUCS90100
[9] FANG W, 1994, P INT C ART NEUR NET, P366
[10] Progress in supervised neural networks
Hush, Don R.
Horne, Bill G.
[J]. IEEE SIGNAL PROCESSING MAGAZINE, 1993, 10 (01) : 8 - 39

← 1 2 3 4 →