Cargando…

Direct training high-performance spiking neural networks for object recognition and detection

INTRODUCTION: The spiking neural network (SNN) is a bionic model that is energy-efficient when implemented on neuromorphic hardwares. The non-differentiability of the spiking signals and the complicated neural dynamics make direct training of high-performance SNNs a great challenge. There are numero...

Descripción completa

Detalles Bibliográficos
Autores principales: Zhang, Hong, Li, Yang, He, Bin, Fan, Xiongfei, Wang, Yue, Zhang, Yu
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Frontiers Media S.A. 2023
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10442545/
https://www.ncbi.nlm.nih.gov/pubmed/37614339
http://dx.doi.org/10.3389/fnins.2023.1229951
_version_ 1785093623919935488
author Zhang, Hong
Li, Yang
He, Bin
Fan, Xiongfei
Wang, Yue
Zhang, Yu
author_facet Zhang, Hong
Li, Yang
He, Bin
Fan, Xiongfei
Wang, Yue
Zhang, Yu
author_sort Zhang, Hong
collection PubMed
description INTRODUCTION: The spiking neural network (SNN) is a bionic model that is energy-efficient when implemented on neuromorphic hardwares. The non-differentiability of the spiking signals and the complicated neural dynamics make direct training of high-performance SNNs a great challenge. There are numerous crucial issues to explore for the deployment of direct training SNNs, such as gradient vanishing and explosion, spiking signal decoding, and applications in upstream tasks. METHODS: To address gradient vanishing, we introduce a binary selection gate into the basic residual block and propose spiking gate (SG) ResNet to implement residual learning in SNNs. We propose two appropriate representations of the gate signal and verify that SG ResNet can overcome gradient vanishing or explosion by analyzing the gradient backpropagation. For the spiking signal decoding, a better decoding scheme than rate coding is achieved by our attention spike decoder (ASD), which dynamically assigns weights to spiking signals along the temporal, channel, and spatial dimensions. RESULTS AND DISCUSSION: The SG ResNet and ASD modules are evaluated on multiple object recognition datasets, including the static ImageNet, CIFAR-100, CIFAR-10, and neuromorphic DVS-CIFAR10 datasets. Superior accuracy is demonstrated with a tiny simulation time step of four, specifically 94.52% top-1 accuracy on CIFAR-10 and 75.64% top-1 accuracy on CIFAR-100. Spiking RetinaNet is proposed using SG ResNet as the backbone and ASD module for information decoding as the first direct-training hybrid SNN-ANN detector for RGB images. Spiking RetinaNet with a SG ResNet34 backbone achieves an mAP of 0.296 on the object detection dataset MSCOCO.
format Online
Article
Text
id pubmed-10442545
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher Frontiers Media S.A.
record_format MEDLINE/PubMed
spelling pubmed-104425452023-08-23 Direct training high-performance spiking neural networks for object recognition and detection Zhang, Hong Li, Yang He, Bin Fan, Xiongfei Wang, Yue Zhang, Yu Front Neurosci Neuroscience INTRODUCTION: The spiking neural network (SNN) is a bionic model that is energy-efficient when implemented on neuromorphic hardwares. The non-differentiability of the spiking signals and the complicated neural dynamics make direct training of high-performance SNNs a great challenge. There are numerous crucial issues to explore for the deployment of direct training SNNs, such as gradient vanishing and explosion, spiking signal decoding, and applications in upstream tasks. METHODS: To address gradient vanishing, we introduce a binary selection gate into the basic residual block and propose spiking gate (SG) ResNet to implement residual learning in SNNs. We propose two appropriate representations of the gate signal and verify that SG ResNet can overcome gradient vanishing or explosion by analyzing the gradient backpropagation. For the spiking signal decoding, a better decoding scheme than rate coding is achieved by our attention spike decoder (ASD), which dynamically assigns weights to spiking signals along the temporal, channel, and spatial dimensions. RESULTS AND DISCUSSION: The SG ResNet and ASD modules are evaluated on multiple object recognition datasets, including the static ImageNet, CIFAR-100, CIFAR-10, and neuromorphic DVS-CIFAR10 datasets. Superior accuracy is demonstrated with a tiny simulation time step of four, specifically 94.52% top-1 accuracy on CIFAR-10 and 75.64% top-1 accuracy on CIFAR-100. Spiking RetinaNet is proposed using SG ResNet as the backbone and ASD module for information decoding as the first direct-training hybrid SNN-ANN detector for RGB images. Spiking RetinaNet with a SG ResNet34 backbone achieves an mAP of 0.296 on the object detection dataset MSCOCO. Frontiers Media S.A. 2023-08-08 /pmc/articles/PMC10442545/ /pubmed/37614339 http://dx.doi.org/10.3389/fnins.2023.1229951 Text en Copyright © 2023 Zhang, Li, He, Fan, Wang and Zhang. https://creativecommons.org/licenses/by/4.0/This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
spellingShingle Neuroscience
Zhang, Hong
Li, Yang
He, Bin
Fan, Xiongfei
Wang, Yue
Zhang, Yu
Direct training high-performance spiking neural networks for object recognition and detection
title Direct training high-performance spiking neural networks for object recognition and detection
title_full Direct training high-performance spiking neural networks for object recognition and detection
title_fullStr Direct training high-performance spiking neural networks for object recognition and detection
title_full_unstemmed Direct training high-performance spiking neural networks for object recognition and detection
title_short Direct training high-performance spiking neural networks for object recognition and detection
title_sort direct training high-performance spiking neural networks for object recognition and detection
topic Neuroscience
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10442545/
https://www.ncbi.nlm.nih.gov/pubmed/37614339
http://dx.doi.org/10.3389/fnins.2023.1229951
work_keys_str_mv AT zhanghong directtraininghighperformancespikingneuralnetworksforobjectrecognitionanddetection
AT liyang directtraininghighperformancespikingneuralnetworksforobjectrecognitionanddetection
AT hebin directtraininghighperformancespikingneuralnetworksforobjectrecognitionanddetection
AT fanxiongfei directtraininghighperformancespikingneuralnetworksforobjectrecognitionanddetection
AT wangyue directtraininghighperformancespikingneuralnetworksforobjectrecognitionanddetection
AT zhangyu directtraininghighperformancespikingneuralnetworksforobjectrecognitionanddetection