A neural network simulator for the Connection Machine

N. Asokan; Ravi Shankar; Kishan Mehrotra; Chilukuri Mohan; Sanjay Ranka

A neural network simulator for the Connection Machine

N. Asokan, Ravi Shankar, Kishan Mehrotra, Chilukuri Mohan, Sanjay Ranka

Department of Electrical Engineering & Computer Science

Research output: Chapter in Book/Entry/Poem › Conference contribution

Abstract

The design, development, and performance of CM-RCS, a neural network simulator for the Connection Machine (CM), are described. The design of the simulator is based on the Rochester Connectionist Simulator (RCS) for connectionist networks. The CM-RCS can be used as a stand-alone system or as a high-performance parallel back end to RCS. In the latter case, once the network has been built by RCS, the back-end system constructs and executes an equivalent network on the CM processor array. The CM-RCS facilitates the exploitation of the massive parallelism inherent in connectionist networks. It also enables substantial reduction in the training times of connectionist networks.

Original language	English (US)
Title of host publication	Proc 5 IEEE Int Symp Intell Control 90
Publisher	IEEE Computer Society
Pages	518-523
Number of pages	6
ISBN (Print)	0818621087
State	Published - 1990
Event	Proceedings of the 5th IEEE International Symposium on Intelligent Control 1990 - Philadelphia, PA, USA Duration: Sep 5 1990 → Sep 7 1990

Publication series

Name	Proc 5 IEEE Int Symp Intell Control 90

Other

Other	Proceedings of the 5th IEEE International Symposium on Intelligent Control 1990
City	Philadelphia, PA, USA
Period	9/5/90 → 9/7/90

ASJC Scopus subject areas

General Engineering

Cite this

@inproceedings{2f58a803a3264bb99b0529532066046b,

title = "A neural network simulator for the Connection Machine",

abstract = "The design, development, and performance of CM-RCS, a neural network simulator for the Connection Machine (CM), are described. The design of the simulator is based on the Rochester Connectionist Simulator (RCS) for connectionist networks. The CM-RCS can be used as a stand-alone system or as a high-performance parallel back end to RCS. In the latter case, once the network has been built by RCS, the back-end system constructs and executes an equivalent network on the CM processor array. The CM-RCS facilitates the exploitation of the massive parallelism inherent in connectionist networks. It also enables substantial reduction in the training times of connectionist networks.",

author = "N. Asokan and Ravi Shankar and Kishan Mehrotra and Chilukuri Mohan and Sanjay Ranka",

year = "1990",

language = "English (US)",

isbn = "0818621087",

series = "Proc 5 IEEE Int Symp Intell Control 90",

publisher = "IEEE Computer Society",

pages = "518--523",

booktitle = "Proc 5 IEEE Int Symp Intell Control 90",

address = "United States",

note = "Proceedings of the 5th IEEE International Symposium on Intelligent Control 1990 ; Conference date: 05-09-1990 Through 07-09-1990",

}

TY - GEN

T1 - A neural network simulator for the Connection Machine

AU - Asokan, N.

AU - Shankar, Ravi

AU - Mehrotra, Kishan

AU - Mohan, Chilukuri

AU - Ranka, Sanjay

PY - 1990

Y1 - 1990

N2 - The design, development, and performance of CM-RCS, a neural network simulator for the Connection Machine (CM), are described. The design of the simulator is based on the Rochester Connectionist Simulator (RCS) for connectionist networks. The CM-RCS can be used as a stand-alone system or as a high-performance parallel back end to RCS. In the latter case, once the network has been built by RCS, the back-end system constructs and executes an equivalent network on the CM processor array. The CM-RCS facilitates the exploitation of the massive parallelism inherent in connectionist networks. It also enables substantial reduction in the training times of connectionist networks.

AB - The design, development, and performance of CM-RCS, a neural network simulator for the Connection Machine (CM), are described. The design of the simulator is based on the Rochester Connectionist Simulator (RCS) for connectionist networks. The CM-RCS can be used as a stand-alone system or as a high-performance parallel back end to RCS. In the latter case, once the network has been built by RCS, the back-end system constructs and executes an equivalent network on the CM processor array. The CM-RCS facilitates the exploitation of the massive parallelism inherent in connectionist networks. It also enables substantial reduction in the training times of connectionist networks.

UR - http://www.scopus.com/inward/record.url?scp=0025530679&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0025530679&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:0025530679

SN - 0818621087

T3 - Proc 5 IEEE Int Symp Intell Control 90

SP - 518

EP - 523

BT - Proc 5 IEEE Int Symp Intell Control 90

PB - IEEE Computer Society

T2 - Proceedings of the 5th IEEE International Symposium on Intelligent Control 1990

Y2 - 5 September 1990 through 7 September 1990

ER -

A neural network simulator for the Connection Machine

Abstract

Publication series

Other

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this