MAXIMUM-ENTROPY, PSEUDOINVERSE TECHNIQUES, AND TIME-SERIES PREDICTIONS WITH LAYERED NETWORKS

被引：7

作者：

DIAMBRA, L

PLASTINO, A

机构：

[1] Departamento de Física, Universidad Nacional de la Plata, (1900) La Plata

来源：

PHYSICAL REVIEW E | 1995年 / 52卷 / 04期

关键词：

D O I：

10.1103/PhysRevE.52.4557

中图分类号：

O35 [流体力学]; O53 [等离子体物理学];

学科分类号：

070204 ; 080103 ; 080704 ;

摘要：

A maximum-entropy-based method for the training of layered networks is presented. Our technique guarantees an errorless learning process for learnable mappings with just a minimum number of examples. The network is proposed for nonlinear systems prediction. Some numerical examples for chaotic time series are presented. The method can be considered to yield an alternative tool for feed-forward training.

引用

页码：4557 / 4560

页数：4

共 18 条

[1] GENERALIZATION ABILITY OF PERCEPTRONS WITH CONTINUOUS OUTPUTS [J].

BOS, S ;

KINZEL, W ;

OPPER, M .

PHYSICAL REVIEW E, 1993, 47 (02) :1384-1391

[2] THE SPACE OF INTERACTIONS IN NEURAL NETWORK MODELS [J].

GARDNER, E .

JOURNAL OF PHYSICS A-MATHEMATICAL AND GENERAL, 1988, 21 (01) :257-270

[3]

HOLLAND JH, 1988, EVOLUTION LEARNING C

[4] INFORMATION THEORY AND STATISTICAL MECHANICS .2. [J].

JAYNES, ET .

PHYSICAL REVIEW, 1957, 108 (02) :171-190

[5] OPTIMIZATION BY SIMULATED ANNEALING [J].

KIRKPATRICK, S ;

GELATT, CD ;

VECCHI, MP .

SCIENCE, 1983, 220 (4598) :671-680

[6]

Lapedes A., 1987, LA UR 87 2662 C 8706

[7]

LAPEDES A, 1987, NEURAL INFORMATION P, P442, DOI DOI 10.1142/9789814434102_0012

[8]

LECUN Y, 1986, DISORDERED SYSTEMS B

[9]

Levine RD, 1978, MAXIMUM ENTROPY PRIN

[10] GENERALIZED LEARNING RULE FOR HIGH-ORDER NEURAL NETWORKS [J].

MATUS, IJ ;

PEREZ, P .

PHYSICAL REVIEW A, 1991, 43 (10) :5683-5688

← 1 2 →