Research map / Architectures and efficiency / Spiking networks
Training Feedback Spiking Neural Networks by Implicit Differentiation on the Equilibrium State
Mingqing Xiao, Qingyan Meng, Zongpeng Zhang, Yisen Wang, Zhouchen Lin
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 |
|---|---|---|---|
| LIF neuron instead of IF neuron | IDE FSNN · MNIST/Fashion-MNIST | best accuracy: 99.55 -> 99.59 | Improved |
| LIF instead of IF | IDE FSNN · N-MNIST | best accuracy: 99.35 -> 99.47 | Improved |
| LIF instead of IF | IDE FSNN · Fashion-MNIST | best accuracy: 90.14 -> 90.25 | Improved |
| increase time steps from 30 to 100 | IDE-LIF AlexNet-F · CIFAR-10 | best accuracy: 91.92 -> 92.15 | Improved |
| increase time steps from 30 to 100 | IDE-LIF CIFARNet-F · CIFAR-10 | best accuracy: 92.23 -> 92.82 | Improved |
| increase time steps from 30 to 100 | IDE-IF CIFARNet-F · CIFAR-100 | best accuracy: 72.10 -> 73.43 | Improved |
| larger CIFARNet-F instead of AlexNet-F at 100 steps | IDE-IF · CIFAR-100 | best accuracy: 72.23 -> 73.43 | Improved |
| IDE implicit differentiation instead of STBP backprop through time | AlexNet-F LIF · CIFAR-10, 30 steps | GPU memory: 11G -> 2.8G | Improved |
| IDE instead of STBP | AlexNet-F LIF · CIFAR-10, 30 steps | accuracy: 87.18 -> 91.74 | Improved |
| increase time steps from 30 to 100 under STBP | STBP AlexNet-F · CIFAR-10 | GPU memory: 11G -> out of memory (approx 36G) | Hurt |
| IDE instead of STBP for firing sparsity | AlexNet-F LIF · CIFAR-10 | total average firing rate: 0.0102 -> 0.0066 | Improved |
| LIF instead of IF for firing sparsity | IDE · CIFAR-10 | total average firing rate: 0.0070 -> 0.0066 | Improved |
Closest papers by meaning
- Take A Shortcut Back: Mitigating the Gradient Vanishing for Training Spiking Neural Networks (NeurIPS 2024 · similarity 0.90)
- Temporal Flexibility in Spiking Neural Networks: Towards Generalization Across Time Steps and Deployment Friendliness (ICLR 2025 · similarity 0.90)
- Enhanced Self-Distillation Framework for Efficient Spiking Neural Network Training (NeurIPS 2025 · similarity 0.89)
- Sign Gradient Descent-based Neuronal Dynamics: ANN-to-SNN Conversion Beyond ReLU Network (ICML 2024 · similarity 0.89)
- Rethinking the Membrane Dynamics and Optimization Objectives of Spiking Neural Networks (NeurIPS 2024 · similarity 0.89)
- NDOT: Neuronal Dynamics-based Online Training for Spiking Neural Networks (ICML 2024 · similarity 0.89)
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.
