Research map / Architectures and efficiency / Spiking networks
Progressive Tandem Learning for Pattern Recognition With Deep Spiking Neural Networks
Jibin Wu, Chenglin Xu, Xiao Jing Han, Daquan Zhou, Malu Zhang, Haizhou Li, Kay Chen Tan
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 |
|---|---|---|---|
| train SNN with progressive tandem learning instead of ANN | AlexNet · Cifar-10 | accuracy: 89.59 -> 90.86 | Improved |
| train SNN with progressive tandem learning instead of ANN | VGG-11 · Cifar-10 | accuracy: 90.59 -> 91.24 | Improved |
| convert to SNN with progressive tandem learning instead of ANN | AlexNet · ImageNet-12 | top-1 accuracy: 58.53 -> 55.19 | Hurt |
| convert to SNN with progressive tandem learning instead of ANN | VGG-16 · ImageNet-12 | top-1 accuracy: 71.65 -> 65.08 | Hurt |
| use progressive tandem learning instead of constrain-then-train (16 vs 200 time steps) | AlexNet SNN · ImageNet-12 | top-1 accuracy: 51.80 -> 55.19 | Improved |
| reduce weight precision from Float32 to 4-bit (ablation) | SNN · Cifar-10 | accuracy: 90.33 -> 89.48 | Hurt |
| reduce weight precision from Float32 to 4-bit (ablation) | SNN · MNIST | accuracy: 99.32 -> 99.29 | No clear effect |
| enlarge encoding time window from 16 to 32 (ablation) | SNN autoencoder · MNIST reconstruction | MSE: 0.01720 -> 0.00662 | Improved |
| enlarge encoding time window from 1 to 16 (ablation) | SNN autoencoder · MNIST reconstruction | MSE: 0.04032 -> 0.01720 | Improved |
| convert ANN separator to SNN | speech separation · WSJ0-2mix open condition | SI-SDR (dB): 12.8 -> 12.2 | Hurt |
| convert ANN separator to SNN | speech separation · WSJ0-2mix open condition | PESQ: 2.94 -> 2.85 | No clear effect |
| use SNN with patience period of only 1 in adaptive scheduler (ablation) | PTL training · Cifar-10 | accuracy | No clear effect |
Closest papers by meaning
- Training Spiking Neural Networks with Local Tandem Learning (NeurIPS 2022 · similarity 0.90)
- Inference-Scale Complexity in ANN-SNN Conversion for High-Performance and Low-Power Applications (CVPR 2025 · similarity 0.89)
- Faster and Stronger: When ANN-SNN Conversion Meets Parallel Spiking Calculation (ICML 2025 · similarity 0.89)
- Adaptive Calibration: A Unified Conversion Framework of Spiking Neural Networks (AAAI 2025 · similarity 0.88)
- Multiplication-Free Parallelizable Spiking Neurons with Efficient Spatio-Temporal Dynamics (NeurIPS 2025 · similarity 0.88)
- Temporal Misalignment in ANN-SNN Conversion and its Mitigation via Probabilistic Spiking Neurons (ICML 2025 · similarity 0.88)
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.
