收藏切换
SpikingReLU: A High-Performance Spiking Neural Network Based on Hybrid Model Conversion
收藏切换
Yufei Wang1, Jingyue Zhao2, Ge Zhang3, Bin Fan4, Zhijie Yang2, Zhong Wan2, Yuchao Dai4, Lei Wang2, *
Journal of Computer Research and Development | 2026, 63(8) : 2143 - 2153
Less
收藏切换
Journal of Computer Research and Development | 2026, 63(8): 2143-2153
• Artificial Intelligence •
SpikingReLU: A High-Performance Spiking Neural Network Based on Hybrid Model Conversion
Full
Yufei Wang1, Jingyue Zhao2, Ge Zhang3, Bin Fan4, Zhijie Yang2, Zhong Wan2, Yuchao Dai4, Lei Wang2, *
Affiliations
  • 1Qiyuan Lab, Beijing 100095
  • 2Institute of National Defense Science and Technology Innovation, Beijing 100091
  • 3Beijing Institute of Tracking and Telecommunication Technology, Beijing 100094
  • 4Shaanxi Key Laboratory of Information Acquisition and Processing (Northwestern Polytechnical University), Xi’an 710129
  • born in 1995. PhD. His main research interests include machine vision and neuromorphic computing

    born in 1994. PhD, assistant professor. Member of CCF. Her main research interests include neuromorphic computing, brain-inspired robotics, and embodied intelligence

    born in 1996. PhD, assistant professor. Her main research interests include hyperspectral image interpretation and multimodal image fusion

    born in 1994. PhD, professor, PhD supervisor. His main research interests include machine vision, computational photography, and 3D reconstruction

    born in 1995. PhD. His main research interests include neuromorphic computing and computer architecture

    born in 1989. PhD, associate professor. Her main research interest includes neuromorphic computing

    born in 1982. PhD, professor, PhD supervisor. Senior member of CCF. His main research interests include machine vision and embodied intelligence

    born in 1977. PhD, professor, PhD supervisor. Distinguished member of CCF. Her main research interests include computer architecture and intelligent chips

About Author:

born in 1995. PhD. His main research interests include machine vision and neuromorphic computing

born in 1994. PhD, assistant professor. Member of CCF. Her main research interests include neuromorphic computing, brain-inspired robotics, and embodied intelligence

born in 1996. PhD, assistant professor. Her main research interests include hyperspectral image interpretation and multimodal image fusion

born in 1994. PhD, professor, PhD supervisor. His main research interests include machine vision, computational photography, and 3D reconstruction

born in 1995. PhD. His main research interests include neuromorphic computing and computer architecture

born in 1989. PhD, associate professor. Her main research interest includes neuromorphic computing

born in 1982. PhD, professor, PhD supervisor. Senior member of CCF. His main research interests include machine vision and embodied intelligence

born in 1977. PhD, professor, PhD supervisor. Distinguished member of CCF. Her main research interests include computer architecture and intelligent chips

Published: 2026-08-01 doi: 10.7544/issn1000-1239.202550722
Outline
收藏切换

Brain-inspired spiking neural networks (SNNs) have shown considerable promise due to their strong spatiotemporal encoding capabilities and event-driven computation. Current mainstream SNN approaches can be divided into two categories: converting pre-trained artificial neural networks (ANNs) into SNNs (ANN conversion methods) and directly training SNNs. However, ANN conversion methods require large time steps to reduce conversion errors, while directly trained SNNs suffer from poor representational capacity. To address these challenges, we propose a high-performance spiking neural network based on hybrid model conversion. The proposed method decouples model training and inference. During the training phase, a portion of the spiking neurons in the SNNs are strategically replaced with ReLU activation functions to enhance the feature learning ability, thereby constructing a high-performance hybrid model that integrates both ANN and SNN components. Since the ANN part of the hybrid model cannot perform event-driven computation, we further introduce a reparameterization technique to convert the entire inference process of the hybrid model into event-driven computation without performance loss. Therefore, the proposed method combines the advantages of high performance and event-driven computation. Experimental results show that the proposed method achieves classification accuracies of 97.31% and 83.34% on CIFAR-10 and CIFAR-100, respectively, 70.89% on ImageNet, and 82.71% on the neuromorphic dataset CIFAR10-DVS, outperforming state-of-the-art methods.

brain-inspired computing  /  spiking neural networks  /  event-driven computing  /  hybrid model  /  training-inference decoupling method
王宇飞, 赵静月, 张歌, 樊斌, 杨智杰, 万众, 戴玉超, 王蕾. SpikingReLU:一种基于混合模型转换的高性能脉冲神经网络. 计算机研究与发展, 2026 , 63 (8) : 2143 -2153 . DOI: 10.7544/issn1000-1239.202550722
Yufei Wang, Jingyue Zhao, Ge Zhang, Bin Fan, Zhijie Yang, Zhong Wan, Yuchao Dai, Lei Wang. SpikingReLU: A High-Performance Spiking Neural Network Based on Hybrid Model Conversion[J]. Journal of Computer Research and Development, 2026 , 63 (8) : 2143 -2153 . DOI: 10.7544/issn1000-1239.202550722
Year 2026 volume 63 Issue 8
74
33
Cite this Article
BibTeX
Article Info
doi: 10.7544/issn1000-1239.202550722
  • Receive Date:2025-09-30
  • Online Date:2026-11-17
  • Published:2026-08-01
Article Data
Affiliations
History
  • Received:2025-09-30
  • Revised:2026-03-30
Affiliations
    1Qiyuan Lab, Beijing 100095
    2Institute of National Defense Science and Technology Innovation, Beijing 100091
    3Beijing Institute of Tracking and Telecommunication Technology, Beijing 100094
    4Shaanxi Key Laboratory of Information Acquisition and Processing (Northwestern Polytechnical University), Xi’an 710129
References
Share
https://castjournals.cast.org.cn/joweb/jsjyjyfz/EN/10.7544/issn1000-1239.202550722
Share to
QR

Scan QR to access full text

Cite this article
BibTeX
Citations
表12种不同金属材料的力学参数
科
Family
属数
Number of
genus
种数
Number of
species
占总种数比例
Percentage of
total species (%)
属
Genus
种数
Number of
species
占总种数比例
Percentage of total
species (%)
鹅膏菌科Amanitaceae 2 11 5.26 鹅膏菌属 Amanita 10 4.78
小菇科 Mycenaceae 2 12 5.74 丝盖伞属 Inocybe 5 2.39
多孔菌科 Polyporaceae 8 14 6.70 蜡蘑属 Laccaria 5 2.39
红菇科 Russulaceae 3 23 11.00 小皮伞属 Marasmius 6 2.87
小菇属 Mycena 11 5.26
光柄菇属 Pluteus 5 2.39
红菇属 Russula 17 8.13
栓菌属 Trametes 5 2.39
关闭全屏
  • BibTeX
  • EndNote
  • RefWorks
  • TxT