Scalable excitatory synaptic circuit design using floating gate based leaky integrators

Vladimir Kornijcuk, Hyungkwang Lim, Inho Kim, Jong Keuk Park, Wook Seong Lee, Jung Hae Choi, Byung Joon Choi, Doo Seok Jeong

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

We propose a scalable synaptic circuit realizing spike timing dependent plasticity (STDP)—compatible with randomly spiking neurons. The feasible working of the circuit was examined by circuit simulation using the BSIM 4.6.0 model. A distinguishable feature of the circuit is the use of floating-gate integrators that provide the compact implementation of biologically plausible relaxation time scale. This relaxation occurs on the basis of charge tunneling that mainly relies upon area-independent tunnel barrier properties (e.g. barrier width and height) rather than capacitance. The circuit simulations feature (i) weight-dependent STDP that spontaneously limits the synaptic weight growth, (ii) competitive synaptic adaptation within both unsupervised and supervised frameworks with randomly spiking neurons. The estimated power consumption is merely 34 pW, perhaps meeting one of the most crucial principles (power-efficiency) of neuromorphic engineering. Finally, a means of fine-tuning the STDP behavior is provided.

Original languageEnglish
Article number17579
JournalScientific reports
Volume7
Issue number1
DOIs
StatePublished - 2017 Dec 1

Fingerprint

Plasticity
Circuit simulation
Neurons
Networks (circuits)
Relaxation time
Tunnels
Electric power utilization
Capacitance
Tuning

Cite this

Kornijcuk, Vladimir ; Lim, Hyungkwang ; Kim, Inho ; Park, Jong Keuk ; Lee, Wook Seong ; Choi, Jung Hae ; Choi, Byung Joon ; Jeong, Doo Seok. / Scalable excitatory synaptic circuit design using floating gate based leaky integrators. In: Scientific reports. 2017 ; Vol. 7, No. 1.
@article{f6cf6b02bb6f44e4aeb400cdb41d4144,
title = "Scalable excitatory synaptic circuit design using floating gate based leaky integrators",
abstract = "We propose a scalable synaptic circuit realizing spike timing dependent plasticity (STDP)—compatible with randomly spiking neurons. The feasible working of the circuit was examined by circuit simulation using the BSIM 4.6.0 model. A distinguishable feature of the circuit is the use of floating-gate integrators that provide the compact implementation of biologically plausible relaxation time scale. This relaxation occurs on the basis of charge tunneling that mainly relies upon area-independent tunnel barrier properties (e.g. barrier width and height) rather than capacitance. The circuit simulations feature (i) weight-dependent STDP that spontaneously limits the synaptic weight growth, (ii) competitive synaptic adaptation within both unsupervised and supervised frameworks with randomly spiking neurons. The estimated power consumption is merely 34 pW, perhaps meeting one of the most crucial principles (power-efficiency) of neuromorphic engineering. Finally, a means of fine-tuning the STDP behavior is provided.",
author = "Vladimir Kornijcuk and Hyungkwang Lim and Inho Kim and Park, {Jong Keuk} and Lee, {Wook Seong} and Choi, {Jung Hae} and Choi, {Byung Joon} and Jeong, {Doo Seok}",
year = "2017",
month = "12",
day = "1",
doi = "10.1038/s41598-017-17889-8",
language = "English",
volume = "7",
journal = "Scientific Reports",
issn = "2045-2322",
number = "1",

}

Scalable excitatory synaptic circuit design using floating gate based leaky integrators. / Kornijcuk, Vladimir; Lim, Hyungkwang; Kim, Inho; Park, Jong Keuk; Lee, Wook Seong; Choi, Jung Hae; Choi, Byung Joon; Jeong, Doo Seok.

In: Scientific reports, Vol. 7, No. 1, 17579, 01.12.2017.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Scalable excitatory synaptic circuit design using floating gate based leaky integrators

AU - Kornijcuk, Vladimir

AU - Lim, Hyungkwang

AU - Kim, Inho

AU - Park, Jong Keuk

AU - Lee, Wook Seong

AU - Choi, Jung Hae

AU - Choi, Byung Joon

AU - Jeong, Doo Seok

PY - 2017/12/1

Y1 - 2017/12/1

N2 - We propose a scalable synaptic circuit realizing spike timing dependent plasticity (STDP)—compatible with randomly spiking neurons. The feasible working of the circuit was examined by circuit simulation using the BSIM 4.6.0 model. A distinguishable feature of the circuit is the use of floating-gate integrators that provide the compact implementation of biologically plausible relaxation time scale. This relaxation occurs on the basis of charge tunneling that mainly relies upon area-independent tunnel barrier properties (e.g. barrier width and height) rather than capacitance. The circuit simulations feature (i) weight-dependent STDP that spontaneously limits the synaptic weight growth, (ii) competitive synaptic adaptation within both unsupervised and supervised frameworks with randomly spiking neurons. The estimated power consumption is merely 34 pW, perhaps meeting one of the most crucial principles (power-efficiency) of neuromorphic engineering. Finally, a means of fine-tuning the STDP behavior is provided.

AB - We propose a scalable synaptic circuit realizing spike timing dependent plasticity (STDP)—compatible with randomly spiking neurons. The feasible working of the circuit was examined by circuit simulation using the BSIM 4.6.0 model. A distinguishable feature of the circuit is the use of floating-gate integrators that provide the compact implementation of biologically plausible relaxation time scale. This relaxation occurs on the basis of charge tunneling that mainly relies upon area-independent tunnel barrier properties (e.g. barrier width and height) rather than capacitance. The circuit simulations feature (i) weight-dependent STDP that spontaneously limits the synaptic weight growth, (ii) competitive synaptic adaptation within both unsupervised and supervised frameworks with randomly spiking neurons. The estimated power consumption is merely 34 pW, perhaps meeting one of the most crucial principles (power-efficiency) of neuromorphic engineering. Finally, a means of fine-tuning the STDP behavior is provided.

UR - http://www.scopus.com/inward/record.url?scp=85054470950&partnerID=8YFLogxK

U2 - 10.1038/s41598-017-17889-8

DO - 10.1038/s41598-017-17889-8

M3 - Article

C2 - 29242504

AN - SCOPUS:85054470950

VL - 7

JO - Scientific Reports

JF - Scientific Reports

SN - 2045-2322

IS - 1

M1 - 17579

ER -