Research map / Architectures and efficiency / Spiking networks
SpikeGrad: An ANN-equivalent Computation Model for Implementing Backpropagation with Spikes
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 spike-discretized gradients instead of full-precision gradients | SpikeGrad with the same topology · MNIST | recognition rate: 99.48 -> 99.52 | No clear effect |
| Use spike-discretized gradients instead of full-precision gradients | SpikeGrad with the same topology · CIFAR10 | recognition rate: 89.72 -> 89.99 | No clear effect |
| Increase error scaling from 50 to 100 (ablation) | SpikeGrad · MNIST | test performance | No clear effect |
| Increase error scaling from 400 to 500 (ablation) | SpikeGrad · CIFAR10 | test performance | No clear effect |
| Use error scaling below approximately 300 (ablation) | SpikeGrad · CIFAR10 | convergence stability: approximately 300 | Hurt |
Closest papers by meaning
- Sparse Spiking Gradient Descent (NeurIPS 2021 · similarity 0.92)
- Efficient Training of Deep Spiking Neural Networks with Input-Driven Derivative-Free Updates (NeurIPS 2026 · similarity 0.90)
- NDOT: Neuronal Dynamics-based Online Training for Spiking Neural Networks (ICML 2024 · similarity 0.89)
- Advancing Training Efficiency of Deep Spiking Neural Networks through Rate-based Backpropagation (NeurIPS 2024 · similarity 0.89)
- Take A Shortcut Back: Mitigating the Gradient Vanishing for Training Spiking Neural Networks (NeurIPS 2024 · similarity 0.89)
- Spike-driven Transformer V2: Meta Spiking Neural Network Architecture Inspiring the Design of Next-generation Neuromorphic Chips (ICLR 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.