Encoding resolution in Hopfield associative memory networks

I. Voitovetsky; H. Guterman

Encoding resolution in Hopfield associative memory networks

Department of Electrical & Computer Engineering

Research output: Contribution to conference › Paper › peer-review

Abstract

The number of precision bits for data storage are limited in the hardware implementation of Hopfield's type networks. This paper addresses the problem of selection of encoding resolution on the synaptic weights. We investigate the effect of quantization in binary and analog Hopfield's networks as a function of data noise and the number of storage patterns. The performance of the networks in terms of recall efficiency was estimated by simulations.

Original language	English
Pages	4.3.2/1-4
State	Published - 1 Jan 1995
Event	Proceedings of the 18th Convention of Electrical and Electronics Engineers in Israel - Tel Aviv, Isr Duration: 7 Mar 1995 → 8 Mar 1995

Conference

Conference	Proceedings of the 18th Convention of Electrical and Electronics Engineers in Israel
City	Tel Aviv, Isr
Period	7/03/95 → 8/03/95

ASJC Scopus subject areas

General Engineering

Cite this

@conference{fdcbcbd2adb848bd903557f740561d91,

title = "Encoding resolution in Hopfield associative memory networks",

abstract = "The number of precision bits for data storage are limited in the hardware implementation of Hopfield's type networks. This paper addresses the problem of selection of encoding resolution on the synaptic weights. We investigate the effect of quantization in binary and analog Hopfield's networks as a function of data noise and the number of storage patterns. The performance of the networks in terms of recall efficiency was estimated by simulations.",

author = "I. Voitovetsky and H. Guterman",

year = "1995",

month = jan,

day = "1",

language = "English",

pages = "4.3.2/1--4",

note = "Proceedings of the 18th Convention of Electrical and Electronics Engineers in Israel ; Conference date: 07-03-1995 Through 08-03-1995",

}

TY - CONF

T1 - Encoding resolution in Hopfield associative memory networks

AU - Voitovetsky, I.

AU - Guterman, H.

PY - 1995/1/1

Y1 - 1995/1/1

N2 - The number of precision bits for data storage are limited in the hardware implementation of Hopfield's type networks. This paper addresses the problem of selection of encoding resolution on the synaptic weights. We investigate the effect of quantization in binary and analog Hopfield's networks as a function of data noise and the number of storage patterns. The performance of the networks in terms of recall efficiency was estimated by simulations.

AB - The number of precision bits for data storage are limited in the hardware implementation of Hopfield's type networks. This paper addresses the problem of selection of encoding resolution on the synaptic weights. We investigate the effect of quantization in binary and analog Hopfield's networks as a function of data noise and the number of storage patterns. The performance of the networks in terms of recall efficiency was estimated by simulations.

UR - http://www.scopus.com/inward/record.url?scp=17944399128&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:17944399128

SP - 4.3.2/1-4

T2 - Proceedings of the 18th Convention of Electrical and Electronics Engineers in Israel

Y2 - 7 March 1995 through 8 March 1995

ER -

Encoding resolution in Hopfield associative memory networks

Abstract

Conference

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this