Multi-input silicon neuron with weighting adaptation

Ming Ze Li; Ping Wang Po; Kea Tiong Tang; Wai-Chi  Fang

doi:10.1109/LISSA.2009.4906744

Multi-input silicon neuron with weighting adaptation

Ming Ze Li^*, Ping Wang Po, Kea Tiong Tang, Wai-Chi Fang

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

This paper presents a biologically inspired "integrate-and-fire (I&F) neuron" which has multiple input dendrites for adaptive weight storage. By using a capacitor-free integrator, longer time constant and smaller chip area can be achieved. A low-power Schmitt Trigger is used to implement the feedback loop to achieve smaller power consumption. Weights are stored by using floating gate MOS transistors as nonvolatile analog memory. Simulation results show that this I&F neuron can be utilized in an analog VLSI neural network system.

Original language	English
Title of host publication	2009 IEEE/NIH Life Science Systems and Applications Workshop, LiSSA 2009
Pages	194-197
Number of pages	4
DOIs	https://doi.org/10.1109/LISSA.2009.4906744
State	Published - 23 Jul 2009
Event	2009 IEEE/NIH Life Science Systems and Applications Workshop, LiSSA 2009 - Bethesda, MD, United States Duration: 9 Apr 2009 → 10 Apr 2009

Publication series

Name	2009 IEEE/NIH Life Science Systems and Applications Workshop, LiSSA 2009

Conference

Conference	2009 IEEE/NIH Life Science Systems and Applications Workshop, LiSSA 2009
Country/Territory	United States
City	Bethesda, MD
Period	9/04/09 → 10/04/09

Access to Document

10.1109/LISSA.2009.4906744

Cite this

@inproceedings{483d8fedaa424f76857c566799c25333,

title = "Multi-input silicon neuron with weighting adaptation",

abstract = "This paper presents a biologically inspired {"}integrate-and-fire (I&F) neuron{"} which has multiple input dendrites for adaptive weight storage. By using a capacitor-free integrator, longer time constant and smaller chip area can be achieved. A low-power Schmitt Trigger is used to implement the feedback loop to achieve smaller power consumption. Weights are stored by using floating gate MOS transistors as nonvolatile analog memory. Simulation results show that this I&F neuron can be utilized in an analog VLSI neural network system.",

author = "Li, {Ming Ze} and Po, {Ping Wang} and Tang, {Kea Tiong} and Wai-Chi Fang",

year = "2009",

month = jul,

day = "23",

doi = "10.1109/LISSA.2009.4906744",

language = "English",

isbn = "9781424442935",

series = "2009 IEEE/NIH Life Science Systems and Applications Workshop, LiSSA 2009",

pages = "194--197",

booktitle = "2009 IEEE/NIH Life Science Systems and Applications Workshop, LiSSA 2009",

note = "2009 IEEE/NIH Life Science Systems and Applications Workshop, LiSSA 2009 ; Conference date: 09-04-2009 Through 10-04-2009",

}

Li, MZ, Po, PW, Tang, KT & Fang, W-C 2009, Multi-input silicon neuron with weighting adaptation. in 2009 IEEE/NIH Life Science Systems and Applications Workshop, LiSSA 2009., 4906744, 2009 IEEE/NIH Life Science Systems and Applications Workshop, LiSSA 2009, pp. 194-197, 2009 IEEE/NIH Life Science Systems and Applications Workshop, LiSSA 2009, Bethesda, MD, United States, 9/04/09. https://doi.org/10.1109/LISSA.2009.4906744

Multi-input silicon neuron with weighting adaptation. / Li, Ming Ze; Po, Ping Wang; Tang, Kea Tiong et al.
2009 IEEE/NIH Life Science Systems and Applications Workshop, LiSSA 2009. 2009. p. 194-197 4906744 (2009 IEEE/NIH Life Science Systems and Applications Workshop, LiSSA 2009).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Multi-input silicon neuron with weighting adaptation

AU - Li, Ming Ze

AU - Po, Ping Wang

AU - Tang, Kea Tiong

AU - Fang, Wai-Chi

PY - 2009/7/23

Y1 - 2009/7/23

N2 - This paper presents a biologically inspired "integrate-and-fire (I&F) neuron" which has multiple input dendrites for adaptive weight storage. By using a capacitor-free integrator, longer time constant and smaller chip area can be achieved. A low-power Schmitt Trigger is used to implement the feedback loop to achieve smaller power consumption. Weights are stored by using floating gate MOS transistors as nonvolatile analog memory. Simulation results show that this I&F neuron can be utilized in an analog VLSI neural network system.

AB - This paper presents a biologically inspired "integrate-and-fire (I&F) neuron" which has multiple input dendrites for adaptive weight storage. By using a capacitor-free integrator, longer time constant and smaller chip area can be achieved. A low-power Schmitt Trigger is used to implement the feedback loop to achieve smaller power consumption. Weights are stored by using floating gate MOS transistors as nonvolatile analog memory. Simulation results show that this I&F neuron can be utilized in an analog VLSI neural network system.

UR - http://www.scopus.com/inward/record.url?scp=67650687128&partnerID=8YFLogxK

U2 - 10.1109/LISSA.2009.4906744

DO - 10.1109/LISSA.2009.4906744

M3 - Conference contribution

AN - SCOPUS:67650687128

SN - 9781424442935

T3 - 2009 IEEE/NIH Life Science Systems and Applications Workshop, LiSSA 2009

SP - 194

EP - 197

BT - 2009 IEEE/NIH Life Science Systems and Applications Workshop, LiSSA 2009

T2 - 2009 IEEE/NIH Life Science Systems and Applications Workshop, LiSSA 2009

Y2 - 9 April 2009 through 10 April 2009

ER -

Multi-input silicon neuron with weighting adaptation

Abstract

Publication series

Conference

Access to Document

Other files and links

Fingerprint

Cite this