Research map / Architectures and efficiency / Spiking networks
TDSNN: From Deep Neural Networks to Deep Spike Neural Networks with Temporal-Coding
Lei Zhang, Shengyuan Zhou, Tian Zhi, Zidong Du, Yunji Chen
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| increase leaky constant L from 1 to 5 (ablation) | TDSNN VGG-16 · ImageNet | top-1/top-5 accuracy: 0%/0% -> 70.87%/90.11% | Improved |
| increase leaky constant L from 1 to 5 (ablation) | TDSNN Alexnet · ImageNet | top-1/top-5 accuracy: 42.08%/66.96% -> 56.7%79.83% | Improved |
| increase leaky constant L from 1 to 5 (ablation) | TDSNN Lenet · MNIST | accuracy: 98.47% -> 99.08% | No clear effect |
| increase leaky constant L from 1 to 5 (ablation) | TDSNN Alexnet · ImageNet | operations reduction vs DNN: 48.4% -> 41.7% | Hurt |
| convert DNN to temporal-coding SNN (TDSNN) vs original DNN | VGG-16 · ImageNet | top-1/top-5 error: 30.82/10.72 -> 29.13/9.89 | Improved |
| convert DNN to temporal-coding SNN (TDSNN) vs original DNN | Lenet · MNIST | error: 0.84 -> 0.92 | Hurt |
| convert DNN to temporal-coding SNN (TDSNN) vs original DNN | Alexnet · ImageNet | top-1/top-5 error: 42.84/19.67 -> 43.3/20.17 | Hurt |
| replace previous SNN conversion with TDSNN | VGG-16 SNN · ImageNet | top-1/top-5 error: 50.39/18.37 -> 29.13/9.89 | Improved |
| replace previous SNN conversion with TDSNN | Lenet SNN · MNIST | error: 0.56 -> 0.92 | Hurt |
| use TDSNN instead of DNN to reduce total operations | Alexnet/VGG-16 · ImageNet | total operations: 41.9%/42.6% reduction | Improved |
| use TDSNN instead of DNN to reduce total operations | Lenet · MNIST | total operations: 1.14x of DNN | Hurt |
Closest papers by meaning
- Temporal-Coded Deep Spiking Neural Network with Easy Training and Robust Performance (AAAI 2021 · similarity 0.89)
- Differential Coding for Training-Free ANN-to-SNN Conversion (ICML 2025 · similarity 0.88)
- TTFSFormer: A TTFS-based Lossless Conversion of Spiking Transformer (ICML 2025 · similarity 0.86)
- Gated Attention Coding for Training High-Performance and Efficient Spiking Neural Networks (AAAI 2024 · similarity 0.85)
- Advancing Training Efficiency of Deep Spiking Neural Networks through Rate-based Backpropagation (NeurIPS 2024 · similarity 0.85)
- TS-SNN: Temporal Shift Module for Spiking Neural Networks (ICML 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
