Linear quantization of Hebbian-type associative memories in interconnection implementations

Pau Choo Chung; Ching Tsorng Tsai; Yung Nien Sun

Linear quantization of Hebbian-type associative memories in interconnection implementations

Pau Choo Chung, Ching Tsorng Tsai, Yung Nien Sun

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

Abstract

The effects of linear quantized Hebbian-type associative memories (HAMs) on storage capacity and hardware implementations are explored in this paper. For the linear quantization, the interconnection weights are linearly quantized into a small number of levels. This consideration focuses mainly on the situation when only a limited accuracy range can be achieved on hardware implementations. Results of simulation and theory show that the number of quantization levels required is relatively small compared with the possible values of interconnections. Therefore, linear quantization in HAMs is worthwhile in hardware implementations.

Original language	English
Pages	1092-1096
Number of pages	5
Publication status	Published - 1994 Dec 1
Event	Proceedings of the 1994 IEEE International Conference on Neural Networks. Part 1 (of 7) - Orlando, FL, USA Duration: 1994 Jun 27 → 1994 Jun 29

Other

Other	Proceedings of the 1994 IEEE International Conference on Neural Networks. Part 1 (of 7)
City	Orlando, FL, USA
Period	94-06-27 → 94-06-29

All Science Journal Classification (ASJC) codes

Software

Fingerprint Dive into the research topics of 'Linear quantization of Hebbian-type associative memories in interconnection implementations'. Together they form a unique fingerprint.

Cite this

@conference{85a8e1b44e2e4f8ea0309747a0558302,

title = "Linear quantization of Hebbian-type associative memories in interconnection implementations",

abstract = "The effects of linear quantized Hebbian-type associative memories (HAMs) on storage capacity and hardware implementations are explored in this paper. For the linear quantization, the interconnection weights are linearly quantized into a small number of levels. This consideration focuses mainly on the situation when only a limited accuracy range can be achieved on hardware implementations. Results of simulation and theory show that the number of quantization levels required is relatively small compared with the possible values of interconnections. Therefore, linear quantization in HAMs is worthwhile in hardware implementations.",

author = "Chung, {Pau Choo} and Tsai, {Ching Tsorng} and Sun, {Yung Nien}",

year = "1994",

month = dec,

day = "1",

language = "English",

pages = "1092--1096",

note = "Proceedings of the 1994 IEEE International Conference on Neural Networks. Part 1 (of 7) ; Conference date: 27-06-1994 Through 29-06-1994",

}

TY - CONF

T1 - Linear quantization of Hebbian-type associative memories in interconnection implementations

AU - Chung, Pau Choo

AU - Tsai, Ching Tsorng

AU - Sun, Yung Nien

PY - 1994/12/1

Y1 - 1994/12/1

N2 - The effects of linear quantized Hebbian-type associative memories (HAMs) on storage capacity and hardware implementations are explored in this paper. For the linear quantization, the interconnection weights are linearly quantized into a small number of levels. This consideration focuses mainly on the situation when only a limited accuracy range can be achieved on hardware implementations. Results of simulation and theory show that the number of quantization levels required is relatively small compared with the possible values of interconnections. Therefore, linear quantization in HAMs is worthwhile in hardware implementations.

AB - The effects of linear quantized Hebbian-type associative memories (HAMs) on storage capacity and hardware implementations are explored in this paper. For the linear quantization, the interconnection weights are linearly quantized into a small number of levels. This consideration focuses mainly on the situation when only a limited accuracy range can be achieved on hardware implementations. Results of simulation and theory show that the number of quantization levels required is relatively small compared with the possible values of interconnections. Therefore, linear quantization in HAMs is worthwhile in hardware implementations.

UR - http://www.scopus.com/inward/record.url?scp=0028744833&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0028744833&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:0028744833

SP - 1092

EP - 1096

T2 - Proceedings of the 1994 IEEE International Conference on Neural Networks. Part 1 (of 7)

Y2 - 27 June 1994 through 29 June 1994

ER -