Using recurrent neural networks to learn the structure of interconnection networks

Mark W. Goudreau, C. Lee Giles

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

A modified Recurrent Neural Network (RNN) is used to learn a Self-Routing Interconnection Network (SRIN) from a set of routing examples. The RNN is modified so that it has several distinct initial states. This is equivalent to a single RNN learning multiple different synchronous sequential machines. We define such a sequential machine structure as augmented and show that a SRIN is essentially an Augmented Synchronous Sequential Machine (ASSM). As an example, we learn a small six-switch SRIN. After training we extract the network's internal representation of the ASSM and corresponding SRIN.

Original languageEnglish (US)
Pages (from-to)793-804
Number of pages12
JournalNeural Networks
Volume8
Issue number5
DOIs
StatePublished - 1995

Fingerprint

Sequential machines
Recurrent neural networks
Learning
Switches

All Science Journal Classification (ASJC) codes

  • Cognitive Neuroscience
  • Artificial Intelligence

Cite this

@article{56f5e2a0e7d14bf58b95d19409a8c3f6,
title = "Using recurrent neural networks to learn the structure of interconnection networks",
abstract = "A modified Recurrent Neural Network (RNN) is used to learn a Self-Routing Interconnection Network (SRIN) from a set of routing examples. The RNN is modified so that it has several distinct initial states. This is equivalent to a single RNN learning multiple different synchronous sequential machines. We define such a sequential machine structure as augmented and show that a SRIN is essentially an Augmented Synchronous Sequential Machine (ASSM). As an example, we learn a small six-switch SRIN. After training we extract the network's internal representation of the ASSM and corresponding SRIN.",
author = "Goudreau, {Mark W.} and Giles, {C. Lee}",
year = "1995",
doi = "10.1016/0893-6080(95)00025-U",
language = "English (US)",
volume = "8",
pages = "793--804",
journal = "Neural Networks",
issn = "0893-6080",
publisher = "Elsevier Limited",
number = "5",

}

Using recurrent neural networks to learn the structure of interconnection networks. / Goudreau, Mark W.; Giles, C. Lee.

In: Neural Networks, Vol. 8, No. 5, 1995, p. 793-804.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Using recurrent neural networks to learn the structure of interconnection networks

AU - Goudreau, Mark W.

AU - Giles, C. Lee

PY - 1995

Y1 - 1995

N2 - A modified Recurrent Neural Network (RNN) is used to learn a Self-Routing Interconnection Network (SRIN) from a set of routing examples. The RNN is modified so that it has several distinct initial states. This is equivalent to a single RNN learning multiple different synchronous sequential machines. We define such a sequential machine structure as augmented and show that a SRIN is essentially an Augmented Synchronous Sequential Machine (ASSM). As an example, we learn a small six-switch SRIN. After training we extract the network's internal representation of the ASSM and corresponding SRIN.

AB - A modified Recurrent Neural Network (RNN) is used to learn a Self-Routing Interconnection Network (SRIN) from a set of routing examples. The RNN is modified so that it has several distinct initial states. This is equivalent to a single RNN learning multiple different synchronous sequential machines. We define such a sequential machine structure as augmented and show that a SRIN is essentially an Augmented Synchronous Sequential Machine (ASSM). As an example, we learn a small six-switch SRIN. After training we extract the network's internal representation of the ASSM and corresponding SRIN.

UR - http://www.scopus.com/inward/record.url?scp=0028852020&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028852020&partnerID=8YFLogxK

U2 - 10.1016/0893-6080(95)00025-U

DO - 10.1016/0893-6080(95)00025-U

M3 - Article

AN - SCOPUS:0028852020

VL - 8

SP - 793

EP - 804

JO - Neural Networks

JF - Neural Networks

SN - 0893-6080

IS - 5

ER -