Research map / Architectures and efficiency / Spiking networks
A Hybrid Neural Coding Approach for Pattern Recognition With Spiking Neural Networks
Xinyi Chen, Yang Qu, Jibin Wu, 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 |
|---|---|---|---|
| use TTFS output coding instead of direct decoding (ablation) | P+B hybrid SNN · SLoClas sound localization | avg inference time: 16 -> 4.37 | Improved |
| use TTFS output coding instead of direct decoding (ablation) | P+R hybrid SNN · SLoClas sound localization | avg inference time: 16 -> 7.62 | Improved |
| use burst hidden coding instead of rate coding (ablation) | P+T hybrid SNN · SLoClas sound localization | MAE (deg): 0.82 -> 0.60 | Improved |
| use burst hidden coding instead of rate coding (ablation) | P+T hybrid SNN · SLoClas sound localization | SynOps (millions): 0.94 -> 2.86 | Hurt |
| use TTFS instead of direct decoding on SynOps (ablation) | P+B hybrid SNN · SLoClas sound localization | SynOps (millions): 10.70 -> 2.86 | Improved |
| raise LTL scaling factor r from 1 to 4 (ablation) | Hybrid coding VGG-16 · CIFAR-10 | accuracy (%): 95.53 -> 91.73 | Hurt |
| raise LTL scaling factor r from 1 to 4 (ablation) | Hybrid coding VGG-16 · CIFAR-100 | accuracy (%): 76.37 -> 64.22 | Hurt |
| raise LTL scaling factor r from 0.2 to 4 (ablation) | Hybrid coding SNN · SLoClas sound localization | accuracy (%): 95.95 -> 96.05 | No clear effect |
| increase burst maximum spike count from 1 to 5 (ablation) | Hybrid coding VGG-16 · CIFAR-100 | accuracy (%): 74.65 -> 76.37 | Improved |
| increase burst maximum spike count from 5 to 10 (ablation) | Hybrid coding VGG-16 · CIFAR-10 | accuracy (%): 95.53 -> 95.43 | No clear effect |
| increase burst maximum spike count from 1 to 5 (ablation) | Hybrid coding SNN · SLoClas sound localization | accuracy (%): 95.22 -> 95.94 | Improved |
| use layer-wise LTL instead of Calibration for D+R+D (ablation) | VGG-16 · CIFAR-10 | accuracy drop from ANN: 1.21% -> 0.56% | Improved |
Closest papers by meaning
- Spiking Neural Networks with Improved Inherent Recurrence Dynamics for Sequential Learning (AAAI 2022 · similarity 0.90)
- A Scalable, Causal, and Energy Efficient Framework for Neural Decoding with Spiking Neural Networks (NeurIPS 2025 · similarity 0.89)
- S$^2$NN: Sub-bit Spiking Neural Networks (NeurIPS 2025 · similarity 0.88)
- EnOF-SNN: Training Accurate Spiking Neural Networks via Enhancing the Output Feature (NeurIPS 2024 · similarity 0.88)
- Self-cross Feature based Spiking Neural Networks for Efficient Few-shot Learning (ICML 2025 · similarity 0.88)
- Spike-driven Transformer V2: Meta Spiking Neural Network Architecture Inspiring the Design of Next-generation Neuromorphic Chips (ICLR 2024 · 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.
