Optimal convergence of on-line backpropagation

被引：39

作者：

Gori, M

Maggini, M

机构：

[1] Dipartimento di Sistemi e Informatica, Università di Firenze, 50139 Firenze, Via di S. Marta

来源：

IEEE TRANSACTIONS ON NEURAL NETWORKS | 1996年 / 7卷 / 01期

关键词：

D O I：

10.1109/72.478415

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Many researchers are quite skeptical about the actual behavior of neural network learning algorithms like backpropagation. One of the major problems is with the lack of clear theoretical results on optimal convergence, particularly for pattern mode algorithms. In this paper, we prove the companion of Rosenblatt's PC (perceptron convergence) theorem for feedforward networks, stating that pattern mode backpropagation converges to an optimal solution for linearly separable patterns.

引用

页码：251 / 254

页数：4

共 16 条

[1]

[Anonymous], 1989, P ADV NEURAL INFORM

[2] NEURAL NETWORKS AND PRINCIPAL COMPONENT ANALYSIS - LEARNING FROM EXAMPLES WITHOUT LOCAL MINIMA [J].

BALDI, P ;

HORNIK, K .

NEURAL NETWORKS, 1989, 2 (01) :53-58

[3] ON THE PROBLEM OF LOCAL MINIMA IN RECURRENT NEURAL NETWORKS [J].

BIANCHINI, M ;

GORI, M ;

MAGGINI, M .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 1994, 5 (02) :167-172

[4]

BIANCHINI M, 1995, MAY IEEE T NEUR NETW, V6, P749

[5] ON THE GEOMETRY OF FEEDFORWARD NEURAL-NETWORK ERROR SURFACES [J].

CHEN, AM ;

LU, HM ;

HECHTNIELSEN, R .

NEURAL COMPUTATION, 1993, 5 (06) :910-927

[6]

Gallant S. I., 1986, Eighth International Conference on Pattern Recognition. Proceedings (Cat. No.86CH2342-4), P849

[7] ON THE PROBLEM OF LOCAL MINIMA IN BACKPROPAGATION [J].

GORI, M ;

TESI, A .

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 1992, 14 (01) :76-86

[8]

JORDAN F, 1991, JUL P INT JOINT C NE, V2, P391

[9]

MINSKY ML, 1988, PERCEPTIONS EXPANDED

[10]

Nilsson Nils., 1965, LEARNING MACHINES

← 1 2 →