skip to main content
Idioma:
Primo Advanced Search
Primo Advanced Search Query Term
Primo Advanced Search prefilters
Busca Simples

CQ ^+ Training: Minimizing Accuracy Loss in Conversion From Convolutional Neural Networks to Spiking Neural Networks

Yan, Zhanglu ; Zhou, Jun ; Wong, Weng-Fai

IEEE transactions on pattern analysis and machine intelligence, 2023-10, Vol.45 (10), p.11600-11611 [Periódico revisado por pares]

United States: IEEE

Texto completo disponível

Citações Citado por
  • Título:
    CQ ^+ Training: Minimizing Accuracy Loss in Conversion From Convolutional Neural Networks to Spiking Neural Networks
  • Autor: Yan, Zhanglu ; Zhou, Jun ; Wong, Weng-Fai
  • Assuntos: Backpropagation ; Clamps ; CNN-to-SNN Conversion ; Convolutional neural networks ; deep spiking neural networks ; Encoding ; Neurons ; Quantization (signal) ; Training
  • É parte de: IEEE transactions on pattern analysis and machine intelligence, 2023-10, Vol.45 (10), p.11600-11611
  • Notas: ObjectType-Article-1
    SourceType-Scholarly Journals-1
    ObjectType-Feature-2
    content type line 23
  • Descrição: Spiking neural networks (SNNs) are attractive for energy-constrained use-cases due to their binarized activation, eliminating the need for weight multiplication. However, its lag in accuracy compared to traditional convolutional network networks (CNNs) has limited its deployment. In this paper, we propose CQ+ training (extended "clamped" and "quantized" training), an SNN-compatible CNN training algorithm that achieves state-of-the-art accuracy for both CIFAR-10 and CIFAR-100 datasets. Using a 7-layer modified VGG model (VGG-*), we achieved 95.06% accuracy on the CIFAR-10 dataset for equivalent SNNs. The accuracy drop from converting the CNN solution to an SNN is only 0.09% when using a time step of 600. To reduce the latency, we propose a parameterized input encoding method and a threshold training method, which further reduces the time window size to 64 while still achieving an accuracy of 94.09%. For the CIFAR-100 dataset, we achieved an accuracy of 77.27% using the same VGG-* structure and a time window of 500. We also demonstrate the transformation of popular CNNs, including ResNet (basic, bottleneck, and shortcut block), MobileNet v1/2, and Densenet, to SNNs with near-zero conversion accuracy loss and a time window size smaller than 60. The framework was developed in PyTorch and is publicly available.
  • Editor: United States: IEEE
  • Idioma: Inglês
Links
Voltar para lista de resultados
Ir para página anteriorAnterior Resultado  6 AvançarIr para próxima página

  • Realização: ABCD-USP USP   Logos de Redes Sociais: Freepik

Buscando em bases de dados remotas. Favor aguardar.