TY - GEN
T1 - Temporal-Coded Deep Spiking Neural Network with Easy Training and Robust Performance
AU - Zhou, Shibo
AU - Li, Xiaohua
AU - Chen, Ying
AU - Chandrasekaran, Sanjeev T.
AU - Sanyal, Arindam
N1 - Publisher Copyright:
Copyright © 2021, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved.
PY - 2021
Y1 - 2021
N2 - Spiking neural network (SNN) is promising but the development has fallen far behind conventional deep neural networks (DNNs) because of difficult training. To resolve the training problem, we analyze the closed-form input-output response of spiking neurons and use the response expression to build abstract SNN models for training. This avoids calculating membrane potential during training and makes the direct training of SNN as efficient as DNN. We show that the nonleaky integrate-and-fire neuron with single-spike temporal-coding is the best choice for direct-train deep SNNs. We develop an energy-efficient phase-domain signal processing circuit for the neuron and propose a direct-train deep SNN framework. Thanks to easy training, we train deep SNNs under weight quantizations to study their robustness over low-cost neuromorphic hardware. Experiments show that our direct-train deep SNNs have the highest CIFAR-10 classification accuracy among SNNs, achieve ImageNet classification accuracy within 1% of the DNN of equivalent architecture, and are robust to weight quantization and noise perturbation.
AB - Spiking neural network (SNN) is promising but the development has fallen far behind conventional deep neural networks (DNNs) because of difficult training. To resolve the training problem, we analyze the closed-form input-output response of spiking neurons and use the response expression to build abstract SNN models for training. This avoids calculating membrane potential during training and makes the direct training of SNN as efficient as DNN. We show that the nonleaky integrate-and-fire neuron with single-spike temporal-coding is the best choice for direct-train deep SNNs. We develop an energy-efficient phase-domain signal processing circuit for the neuron and propose a direct-train deep SNN framework. Thanks to easy training, we train deep SNNs under weight quantizations to study their robustness over low-cost neuromorphic hardware. Experiments show that our direct-train deep SNNs have the highest CIFAR-10 classification accuracy among SNNs, achieve ImageNet classification accuracy within 1% of the DNN of equivalent architecture, and are robust to weight quantization and noise perturbation.
UR - http://www.scopus.com/inward/record.url?scp=85108543030&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85108543030&partnerID=8YFLogxK
M3 - Conference contribution
AN - SCOPUS:85108543030
T3 - 35th AAAI Conference on Artificial Intelligence, AAAI 2021
SP - 11143
EP - 11151
BT - 35th AAAI Conference on Artificial Intelligence, AAAI 2021
PB - Association for the Advancement of Artificial Intelligence
T2 - 35th AAAI Conference on Artificial Intelligence, AAAI 2021
Y2 - 2 February 2021 through 9 February 2021
ER -