Biomimetic, soft-material synapse for neuromorphic computing: From device to network

Md Sakib Hasan, Catherine D. Schuman, Joseph Najem, Ryan Weiss, Nicholas D. Skuda, Alex Belianinov, C. Patrick Collier, Stephen A. Sarles, Garrett S. Rose

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Neuromorphic computing refers to a variety of brain-inspired computers, devices, and models inspired by the interconnectivity, performance, and energy efficiency of the human brain. Unlike the ubiquitous von Neumann computer architectures with complex processor cores and sequential computation, biological neurons and synapses operate by storing and processing information simultaneously with the capacity of flexible adaptation resulting in massive computational capability with much less power consumption. The search for a synaptic material which can closely imitate bio-synapse has led to an alamethicin-doped, synthetic biomembrane which can emulate key synaptic functions due to generic memristive property enabling learning and computation. This two-terminal, biomolecular memristor, in contrast to its solid-state counterparts, features similar structure, switching mechanism, and ionic transport modality as biological synapses while consuming considerably lower power. In this paper, we outline a methodology for using this biomolecular synapse to build neural networks capable of solving real-world problems. The physical mechanism underlying its volatile memristance is explored followed by the development of a model of this device for circuit simulation. We outline a circuit design technique to integrate this synapse with solid-state neuron circuit for hardware implementation. Based on these results, we develop a high level simulation framework and use a training scheme called Evolutionary Optimization for Neuromorphic System (EONS) to generate networks for solving two problems, namely iris dataset classification and EEG classification task. The small network size and comparable to state-of-the-art accuracy of these preliminary networks show its potential to enhance synaptic functionality in next generation neuromorphic hardware.

Original languageEnglish (US)
Title of host publicationProceedings of the 2018 IEEE Dallas Circuits and Systems Conference, DCAS 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781538692622
DOIs
StatePublished - Jan 18 2019
Event13th IEEE Dallas Circuits and Systems Conference, DCAS 2018 - Richardson, United States
Duration: Nov 12 2018 → …

Publication series

NameProceedings of the 2018 IEEE Dallas Circuits and Systems Conference, DCAS 2018

Conference

Conference13th IEEE Dallas Circuits and Systems Conference, DCAS 2018
CountryUnited States
CityRichardson
Period11/12/18 → …

Fingerprint

Biomimetics
Neurons
Brain
Memristors
Hardware
Computer architecture
Networks (circuits)
Circuit simulation
Electroencephalography
Energy efficiency
Electric power utilization
Neural networks

All Science Journal Classification (ASJC) codes

  • Electrical and Electronic Engineering
  • Hardware and Architecture

Cite this

Hasan, M. S., Schuman, C. D., Najem, J., Weiss, R., Skuda, N. D., Belianinov, A., ... Rose, G. S. (2019). Biomimetic, soft-material synapse for neuromorphic computing: From device to network. In Proceedings of the 2018 IEEE Dallas Circuits and Systems Conference, DCAS 2018 [8620187] (Proceedings of the 2018 IEEE Dallas Circuits and Systems Conference, DCAS 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/DCAS.2018.8620187
Hasan, Md Sakib ; Schuman, Catherine D. ; Najem, Joseph ; Weiss, Ryan ; Skuda, Nicholas D. ; Belianinov, Alex ; Collier, C. Patrick ; Sarles, Stephen A. ; Rose, Garrett S. / Biomimetic, soft-material synapse for neuromorphic computing : From device to network. Proceedings of the 2018 IEEE Dallas Circuits and Systems Conference, DCAS 2018. Institute of Electrical and Electronics Engineers Inc., 2019. (Proceedings of the 2018 IEEE Dallas Circuits and Systems Conference, DCAS 2018).
@inproceedings{9c648951f6e6437fab93dab93f3c67fe,
title = "Biomimetic, soft-material synapse for neuromorphic computing: From device to network",
abstract = "Neuromorphic computing refers to a variety of brain-inspired computers, devices, and models inspired by the interconnectivity, performance, and energy efficiency of the human brain. Unlike the ubiquitous von Neumann computer architectures with complex processor cores and sequential computation, biological neurons and synapses operate by storing and processing information simultaneously with the capacity of flexible adaptation resulting in massive computational capability with much less power consumption. The search for a synaptic material which can closely imitate bio-synapse has led to an alamethicin-doped, synthetic biomembrane which can emulate key synaptic functions due to generic memristive property enabling learning and computation. This two-terminal, biomolecular memristor, in contrast to its solid-state counterparts, features similar structure, switching mechanism, and ionic transport modality as biological synapses while consuming considerably lower power. In this paper, we outline a methodology for using this biomolecular synapse to build neural networks capable of solving real-world problems. The physical mechanism underlying its volatile memristance is explored followed by the development of a model of this device for circuit simulation. We outline a circuit design technique to integrate this synapse with solid-state neuron circuit for hardware implementation. Based on these results, we develop a high level simulation framework and use a training scheme called Evolutionary Optimization for Neuromorphic System (EONS) to generate networks for solving two problems, namely iris dataset classification and EEG classification task. The small network size and comparable to state-of-the-art accuracy of these preliminary networks show its potential to enhance synaptic functionality in next generation neuromorphic hardware.",
author = "Hasan, {Md Sakib} and Schuman, {Catherine D.} and Joseph Najem and Ryan Weiss and Skuda, {Nicholas D.} and Alex Belianinov and Collier, {C. Patrick} and Sarles, {Stephen A.} and Rose, {Garrett S.}",
year = "2019",
month = "1",
day = "18",
doi = "10.1109/DCAS.2018.8620187",
language = "English (US)",
series = "Proceedings of the 2018 IEEE Dallas Circuits and Systems Conference, DCAS 2018",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
booktitle = "Proceedings of the 2018 IEEE Dallas Circuits and Systems Conference, DCAS 2018",
address = "United States",

}

Hasan, MS, Schuman, CD, Najem, J, Weiss, R, Skuda, ND, Belianinov, A, Collier, CP, Sarles, SA & Rose, GS 2019, Biomimetic, soft-material synapse for neuromorphic computing: From device to network. in Proceedings of the 2018 IEEE Dallas Circuits and Systems Conference, DCAS 2018., 8620187, Proceedings of the 2018 IEEE Dallas Circuits and Systems Conference, DCAS 2018, Institute of Electrical and Electronics Engineers Inc., 13th IEEE Dallas Circuits and Systems Conference, DCAS 2018, Richardson, United States, 11/12/18. https://doi.org/10.1109/DCAS.2018.8620187

Biomimetic, soft-material synapse for neuromorphic computing : From device to network. / Hasan, Md Sakib; Schuman, Catherine D.; Najem, Joseph; Weiss, Ryan; Skuda, Nicholas D.; Belianinov, Alex; Collier, C. Patrick; Sarles, Stephen A.; Rose, Garrett S.

Proceedings of the 2018 IEEE Dallas Circuits and Systems Conference, DCAS 2018. Institute of Electrical and Electronics Engineers Inc., 2019. 8620187 (Proceedings of the 2018 IEEE Dallas Circuits and Systems Conference, DCAS 2018).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Biomimetic, soft-material synapse for neuromorphic computing

T2 - From device to network

AU - Hasan, Md Sakib

AU - Schuman, Catherine D.

AU - Najem, Joseph

AU - Weiss, Ryan

AU - Skuda, Nicholas D.

AU - Belianinov, Alex

AU - Collier, C. Patrick

AU - Sarles, Stephen A.

AU - Rose, Garrett S.

PY - 2019/1/18

Y1 - 2019/1/18

N2 - Neuromorphic computing refers to a variety of brain-inspired computers, devices, and models inspired by the interconnectivity, performance, and energy efficiency of the human brain. Unlike the ubiquitous von Neumann computer architectures with complex processor cores and sequential computation, biological neurons and synapses operate by storing and processing information simultaneously with the capacity of flexible adaptation resulting in massive computational capability with much less power consumption. The search for a synaptic material which can closely imitate bio-synapse has led to an alamethicin-doped, synthetic biomembrane which can emulate key synaptic functions due to generic memristive property enabling learning and computation. This two-terminal, biomolecular memristor, in contrast to its solid-state counterparts, features similar structure, switching mechanism, and ionic transport modality as biological synapses while consuming considerably lower power. In this paper, we outline a methodology for using this biomolecular synapse to build neural networks capable of solving real-world problems. The physical mechanism underlying its volatile memristance is explored followed by the development of a model of this device for circuit simulation. We outline a circuit design technique to integrate this synapse with solid-state neuron circuit for hardware implementation. Based on these results, we develop a high level simulation framework and use a training scheme called Evolutionary Optimization for Neuromorphic System (EONS) to generate networks for solving two problems, namely iris dataset classification and EEG classification task. The small network size and comparable to state-of-the-art accuracy of these preliminary networks show its potential to enhance synaptic functionality in next generation neuromorphic hardware.

AB - Neuromorphic computing refers to a variety of brain-inspired computers, devices, and models inspired by the interconnectivity, performance, and energy efficiency of the human brain. Unlike the ubiquitous von Neumann computer architectures with complex processor cores and sequential computation, biological neurons and synapses operate by storing and processing information simultaneously with the capacity of flexible adaptation resulting in massive computational capability with much less power consumption. The search for a synaptic material which can closely imitate bio-synapse has led to an alamethicin-doped, synthetic biomembrane which can emulate key synaptic functions due to generic memristive property enabling learning and computation. This two-terminal, biomolecular memristor, in contrast to its solid-state counterparts, features similar structure, switching mechanism, and ionic transport modality as biological synapses while consuming considerably lower power. In this paper, we outline a methodology for using this biomolecular synapse to build neural networks capable of solving real-world problems. The physical mechanism underlying its volatile memristance is explored followed by the development of a model of this device for circuit simulation. We outline a circuit design technique to integrate this synapse with solid-state neuron circuit for hardware implementation. Based on these results, we develop a high level simulation framework and use a training scheme called Evolutionary Optimization for Neuromorphic System (EONS) to generate networks for solving two problems, namely iris dataset classification and EEG classification task. The small network size and comparable to state-of-the-art accuracy of these preliminary networks show its potential to enhance synaptic functionality in next generation neuromorphic hardware.

UR - http://www.scopus.com/inward/record.url?scp=85062427936&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85062427936&partnerID=8YFLogxK

U2 - 10.1109/DCAS.2018.8620187

DO - 10.1109/DCAS.2018.8620187

M3 - Conference contribution

AN - SCOPUS:85062427936

T3 - Proceedings of the 2018 IEEE Dallas Circuits and Systems Conference, DCAS 2018

BT - Proceedings of the 2018 IEEE Dallas Circuits and Systems Conference, DCAS 2018

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Hasan MS, Schuman CD, Najem J, Weiss R, Skuda ND, Belianinov A et al. Biomimetic, soft-material synapse for neuromorphic computing: From device to network. In Proceedings of the 2018 IEEE Dallas Circuits and Systems Conference, DCAS 2018. Institute of Electrical and Electronics Engineers Inc. 2019. 8620187. (Proceedings of the 2018 IEEE Dallas Circuits and Systems Conference, DCAS 2018). https://doi.org/10.1109/DCAS.2018.8620187