Research map / Architectures and efficiency / Spiking networks
Stabilizing Spiking Neurons Through Biologically Inspired Polarization
Matthew Lai, Longbing Cao
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 POLARA neuron instead of LIF with tuned surrogate gradient | 1-layer SNN · MNIST | accuracy: 96.39% -> 97.49% | Improved |
| use POLARA neuron instead of LIF | 1-layer SNN · Fashion-MNIST | accuracy: 85.00% -> 88.78% | Improved |
| use POLARA neuron instead of LIF | 1-layer SNN · EMNIST-byclass | accuracy: 83.18% -> 85.12% | Improved |
| POLARA with ResNet-38 and 50 time steps vs best short-horizon SNN (AGL) | POLARA · CIFAR-10 | accuracy: 96.18% (AGL, 2 steps) -> 94.00% | Hurt |
| POLARA vs GLIF on CIFAR-100 | POLARA · CIFAR-100 | accuracy: 77.28% (GLIF) -> 75.12% | Hurt |
| add membrane decay and stochastic amplification (ablation) | POLARA · MNIST | accuracy: 45.18% -> 97.49% | Improved |
| add shaped spike activation instead of step function with surrogate gradient (ablation) | POLARA · MNIST | accuracy: 96.80% -> 97.49% | Improved |
| add refractory suppression (ablation) | POLARA · MNIST | accuracy: 94.24% -> 97.49% | Improved |
| add adaptive thresholding instead of fixed threshold (ablation) | POLARA · MNIST | accuracy: 94.25% -> 97.49% | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- GLIF — GLIF: A Unified Gated Leaky Integrate-and-Fire Neuron for Spiking Neural Networks (NeurIPS 2022) · better on 0 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Neuronal Self-Adaptation Enhances Capacity and Robustness of Representation in Spiking Neural Networks (NeurIPS 2026 · similarity 0.87)
- Rethinking the Membrane Dynamics and Optimization Objectives of Spiking Neural Networks (NeurIPS 2024 · similarity 0.87)
- The Dormant Spiking Neuron: A State-Driven Mechanism for Efficient Spiking Neural Networks (NeurIPS 2026 · similarity 0.86)
- P-SPIKESSM: HARNESSING PROBABILISTIC SPIKING STATE SPACE MODELS FOR LONG-RANGE DEPENDENCY TASKS (ICLR 2025 · similarity 0.86)
- Balanced Resonate-and-Fire Neurons (ICML 2024 · similarity 0.86)
- Robust Stable Spiking Neural Networks (ICML 2024 · 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.
