PARALLEL BACKPROPAGATION LEARNING ALGORITHMS ON CRAY Y-MP8/864 SUPERCOMPUTER

被引：119

作者：

HUNG, SL ^{[1
]}

ADELI, H ^{[1
]}

机构：

[1] OHIO STATE UNIV,COLL ENGN,COLUMBUS,OH 43210

来源：

NEUROCOMPUTING | 1993年 / 5卷 / 06期

关键词：

BACKPROPAGATION; MULTITASKING; NEURAL NETWORKS; PARALLEL PROCESSING; VECTORIZATION;

D O I：

10.1016/0925-2312(93)90042-2

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Parallel backpropagation neural networks learning algorithms have been developed employing the vectorization and microtasking capabilities of vector MIMD machines. They have been implemented in C on CRAY Y-MP/864 supercomputer under UNICOS operating system. The algorithms have been applied to two different domains: engineering design and image recognition, and their performance has been investigated. A maximum speedup of about 6.7 is achieved using eight processors for a large network with 5950 links due to microtasking only. When vectorization is combined with microtasking, a maximum speedup of about 33 is realized using eight processors.

引用

页码：287 / 302

页数：16

共 10 条

[1]

Adeli H., 1989, Microcomputers in Civil Engineering, V4, P247

[2]

Adeli H, 1992, PARALLEL PROCESSING, P1

[3]

Hung S. L., 1991, Neurocomputing, V3, P3, DOI 10.1016/0925-2312(91)90016-5

[4]

McClelland I.L., 1986, PARALLEL DISTRIBUTED, V2, P170

[5]

ROSENBERG CR, 1988, CONNECTIONIST MODELS, P329

[6]

Rumelhart David E., 1987, LEARNING INTERNAL RE, P318

[7]

SALEH A, 1993, J AEROSPACE ENG, V6

[8]

1990, CRAYYMP CRAY XMP EA

[9]

1991, CRAY STANDARD C PROG

[10]

1986, MANUAL STEEL CONSTRU

← 1 →