Research map / Architectures and efficiency / Spiking networks
Robust Selective Activation with Randomized Temporal K-Winner-Take-All in Spiking Neural Networks for Continual Learning
Jiangrong Shen, Liang Zhao, Qi Xu, Yuqi Yang, Liangjun Chen, Gang Pan, Badong 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 |
|---|---|---|---|
| add probabilistic selection (alpha>0) (ablation) | RTK-WTA + EWC · splitCIFAR100 | accuracy: 36.47 -> 41.46 | Improved |
| add probabilistic selection (alpha>0) (ablation) | RTK-WTA + EWC · splitMNIST | accuracy: 82.18 -> 85.25 | Improved |
| add probabilistic selection (alpha>0) (ablation) | RTK-WTA + EWC · splitCIFAR10 | accuracy: 80.09 -> 80.37 | No clear effect |
| add trace-based neuron dynamics (ablation) | RTK-WTA + EWC · splitCIFAR100 | accuracy: 22.21 -> 41.46 | Improved |
| add trace-based neuron dynamics (ablation) | RTK-WTA + EWC · splitMNIST | accuracy: 80.12 -> 85.25 | Improved |
| add randomized selection to trace-based K-WTA | Trace-based SA-SNN · splitCIFAR100 | accuracy: 22.86 -> 32.91 | Improved |
| add randomized selection to trace-based K-WTA | Trace-based SA-SNN · splitMNIST | accuracy: 60.06 -> 60.37 | No clear effect |
| add EWC | RTK-WTA · splitCIFAR100 | accuracy: 32.91 -> 41.46 | Improved |
| add randomness to rate-based K-WTA | Rate-based SA-SNN · splitCIFAR100 | accuracy: 21.37 -> 20.76 | No clear effect |
| raise randomness coefficient alpha from 0.1 to 0.5 (ablation) | Random-k step · accuracy | accuracy: -14.47% drop | Hurt |
| add RTK-WTA to SNN-ViT | Trace-based SA-SNN · Tiny ImageNet | accuracy: 23.22 -> 26.42 | Improved |
| add RTK-WTA noise robustness | Trace-based SA-SNN · Tiny ImageNet noise 0.5 | accuracy: 22.03 -> 35.07 | Improved |
Closest papers by meaning
- Efficient Spiking Neural Networks with Sparse Selective Activation for Continual Learning (AAAI 2024 · similarity 0.92)
- Temporal Flexibility in Spiking Neural Networks: Towards Generalization Across Time Steps and Deployment Friendliness (ICLR 2025 · similarity 0.86)
- Learning the Plasticity: Plasticity-Driven Learning Framework in Spiking Neural Networks (NeurIPS 2025 · similarity 0.86)
- SpikingSSMs: Learning Long Sequences with Sparse and Parallel Spiking State Space Models (AAAI 2025 · similarity 0.86)
- When to Learn and When to Stop: Quitting at the Optimal Time (Student Abstract) (AAAI 2025 · similarity 0.86)
- The Dormant Spiking Neuron: A State-Driven Mechanism for Efficient Spiking Neural Networks (NeurIPS 2026 · similarity 0.86)
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.
