Unifying algorithm/architecture for artificial neural networks

S. Y. Kung; J. N. Hwang

Unifying algorithm/architecture for artificial neural networks

Research output: Contribution to journal › Conference article › peer-review

Abstract

A generic iterative model is presented for a wide variety of artificial neural networks (ANNs): single-layer feedback networks, multilayer feed-forward networks, hierarchical competitive networks, and hidden Markov models. Unifying mathematical formulations are provided for both the retrieving and learning phases of ANNs. Based on the unifying mathematical formulation, a programmable universal ring systolic array is derived for both phases. It maximizes the strength of VLSI in terms of intensive and pipelined computing and yet circumvents the limitation on communication. Hardware implementation for the processing units based on CORDIC techniques is discussed.

Original language	English (US)
Pages (from-to)	2505-2508
Number of pages	4
Journal	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume	4
State	Published - 1989
Event	1989 International Conference on Acoustics, Speech, and Signal Processing - Glasgow, Scotland Duration: May 23 1989 → May 26 1989

All Science Journal Classification (ASJC) codes

Software
Signal Processing
Electrical and Electronic Engineering

Cite this

@article{165164db67c7499193974f0e90149aca,

title = "Unifying algorithm/architecture for artificial neural networks",

abstract = "A generic iterative model is presented for a wide variety of artificial neural networks (ANNs): single-layer feedback networks, multilayer feed-forward networks, hierarchical competitive networks, and hidden Markov models. Unifying mathematical formulations are provided for both the retrieving and learning phases of ANNs. Based on the unifying mathematical formulation, a programmable universal ring systolic array is derived for both phases. It maximizes the strength of VLSI in terms of intensive and pipelined computing and yet circumvents the limitation on communication. Hardware implementation for the processing units based on CORDIC techniques is discussed.",

author = "Kung, {S. Y.} and Hwang, {J. N.}",

year = "1989",

language = "English (US)",

volume = "4",

pages = "2505--2508",

journal = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

issn = "0736-7791",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

note = "1989 International Conference on Acoustics, Speech, and Signal Processing ; Conference date: 23-05-1989 Through 26-05-1989",

}

TY - JOUR

T1 - Unifying algorithm/architecture for artificial neural networks

AU - Kung, S. Y.

AU - Hwang, J. N.

PY - 1989

Y1 - 1989

N2 - A generic iterative model is presented for a wide variety of artificial neural networks (ANNs): single-layer feedback networks, multilayer feed-forward networks, hierarchical competitive networks, and hidden Markov models. Unifying mathematical formulations are provided for both the retrieving and learning phases of ANNs. Based on the unifying mathematical formulation, a programmable universal ring systolic array is derived for both phases. It maximizes the strength of VLSI in terms of intensive and pipelined computing and yet circumvents the limitation on communication. Hardware implementation for the processing units based on CORDIC techniques is discussed.

AB - A generic iterative model is presented for a wide variety of artificial neural networks (ANNs): single-layer feedback networks, multilayer feed-forward networks, hierarchical competitive networks, and hidden Markov models. Unifying mathematical formulations are provided for both the retrieving and learning phases of ANNs. Based on the unifying mathematical formulation, a programmable universal ring systolic array is derived for both phases. It maximizes the strength of VLSI in terms of intensive and pipelined computing and yet circumvents the limitation on communication. Hardware implementation for the processing units based on CORDIC techniques is discussed.

UR - http://www.scopus.com/inward/record.url?scp=0024928793&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0024928793&partnerID=8YFLogxK

M3 - Conference article

AN - SCOPUS:0024928793

SN - 0736-7791

VL - 4

SP - 2505

EP - 2508

JO - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

JF - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

T2 - 1989 International Conference on Acoustics, Speech, and Signal Processing

Y2 - 23 May 1989 through 26 May 1989

ER -

Unifying algorithm/architecture for artificial neural networks

Abstract

All Science Journal Classification (ASJC) codes

Other files and links

Fingerprint

Cite this