Robustness to Noisy Synaptic Weights in Spiking Neural NetworksCitation formats

Standard

Robustness to Noisy Synaptic Weights in Spiking Neural Networks. / Li, Chen; Chen, Runze; Moutafis, Christoforos; Furber, Steve.

2020 International Joint Conference on Neural Networks (IJCNN). Glasgow, United Kingdom, United Kingdom : IEEE, 2020. p. 1-8.

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Harvard

Li, C, Chen, R, Moutafis, C & Furber, S 2020, Robustness to Noisy Synaptic Weights in Spiking Neural Networks. in 2020 International Joint Conference on Neural Networks (IJCNN). IEEE, Glasgow, United Kingdom, United Kingdom, pp. 1-8. https://doi.org/10.1109/IJCNN48605.2020.9207019

APA

Li, C., Chen, R., Moutafis, C., & Furber, S. (2020). Robustness to Noisy Synaptic Weights in Spiking Neural Networks. In 2020 International Joint Conference on Neural Networks (IJCNN) (pp. 1-8). IEEE. https://doi.org/10.1109/IJCNN48605.2020.9207019

Vancouver

Li C, Chen R, Moutafis C, Furber S. Robustness to Noisy Synaptic Weights in Spiking Neural Networks. In 2020 International Joint Conference on Neural Networks (IJCNN). Glasgow, United Kingdom, United Kingdom: IEEE. 2020. p. 1-8 https://doi.org/10.1109/IJCNN48605.2020.9207019

Author

Li, Chen ; Chen, Runze ; Moutafis, Christoforos ; Furber, Steve. / Robustness to Noisy Synaptic Weights in Spiking Neural Networks. 2020 International Joint Conference on Neural Networks (IJCNN). Glasgow, United Kingdom, United Kingdom : IEEE, 2020. pp. 1-8

Bibtex

@inproceedings{dd12598a3fce4a54a4b638634c4162ac,
title = "Robustness to Noisy Synaptic Weights in Spiking Neural Networks",
abstract = "Spiking neural networks (SNNs) are promising neural network models to achieve power-efficient and event-based computing on neuromorphic hardware. SNNs inherently contain noise and are robust to noisy inputs as well as noise related to the discrete 1-bit spike. In this paper, we find that SNNs are more robust to Gaussian noise in synaptic weights than artificial neural networks (ANNs) under some conditions. This finding will enhance our understanding of the neural dynamics in SNNs and of the advantages of SNNs compared with ANNs. Our results imply the possibility of using high-performance cutting-edge materials with intrinsic noise as an information storage medium in SNNs.",
keywords = "Spiking Neural Network, Artificial Neural Network, noisy weights, Gaussian noise",
author = "Chen Li and Runze Chen and Christoforos Moutafis and Steve Furber",
year = "2020",
month = sep,
day = "28",
doi = "10.1109/IJCNN48605.2020.9207019",
language = "English",
isbn = "978-1-7281-6927-9",
pages = "1--8",
booktitle = "2020 International Joint Conference on Neural Networks (IJCNN)",
publisher = "IEEE",
address = "United States",

}

RIS

TY - GEN

T1 - Robustness to Noisy Synaptic Weights in Spiking Neural Networks

AU - Li, Chen

AU - Chen, Runze

AU - Moutafis, Christoforos

AU - Furber, Steve

PY - 2020/9/28

Y1 - 2020/9/28

N2 - Spiking neural networks (SNNs) are promising neural network models to achieve power-efficient and event-based computing on neuromorphic hardware. SNNs inherently contain noise and are robust to noisy inputs as well as noise related to the discrete 1-bit spike. In this paper, we find that SNNs are more robust to Gaussian noise in synaptic weights than artificial neural networks (ANNs) under some conditions. This finding will enhance our understanding of the neural dynamics in SNNs and of the advantages of SNNs compared with ANNs. Our results imply the possibility of using high-performance cutting-edge materials with intrinsic noise as an information storage medium in SNNs.

AB - Spiking neural networks (SNNs) are promising neural network models to achieve power-efficient and event-based computing on neuromorphic hardware. SNNs inherently contain noise and are robust to noisy inputs as well as noise related to the discrete 1-bit spike. In this paper, we find that SNNs are more robust to Gaussian noise in synaptic weights than artificial neural networks (ANNs) under some conditions. This finding will enhance our understanding of the neural dynamics in SNNs and of the advantages of SNNs compared with ANNs. Our results imply the possibility of using high-performance cutting-edge materials with intrinsic noise as an information storage medium in SNNs.

KW - Spiking Neural Network

KW - Artificial Neural Network

KW - noisy weights

KW - Gaussian noise

U2 - 10.1109/IJCNN48605.2020.9207019

DO - 10.1109/IJCNN48605.2020.9207019

M3 - Conference contribution

SN - 978-1-7281-6927-9

SP - 1

EP - 8

BT - 2020 International Joint Conference on Neural Networks (IJCNN)

PB - IEEE

CY - Glasgow, United Kingdom, United Kingdom

ER -