Research map / Optimization / Optimizers
Gradient Descent for Spiking Neural Networks
Dongsung Huh, Terrence J. Sejnowski
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 |
|---|---|---|---|
| Train a spiking network with the differentiable synapse model and exact gradient calculation | 30 NIF neurons · Auto-encoder task | output tracking | Improved |
| Train a spiking network with the differentiable synapse model and exact gradient calculation | 30 NIF neurons · General predictive coding task | output tracking | Improved |
| Train a spiking network with the differentiable synapse model and exact gradient calculation | 80 QIF neurons · Delayed-memory XOR task | task performance | Improved |
| Use NIF neurons instead of QIF neurons (ablation) | Spiking network trained on delayed-memory XOR · Delayed-memory XOR task | task performance | Hurt |
| Set regular recurrent weights W to zero (ablation) | Spiking network trained on auto-encoding · Auto-encoder task | task performance | No clear effect |
| Use finite-time-constant fast synapses rather than instantaneously fast synapses | Predictive-coding spiking networks · Predictive coding tasks | efficient coding | No clear effect |
| Use intrinsic QIF spiking nonlinearity without extra dendritic nonlinearity | Spiking network trained on delayed-memory XOR · Delayed-memory XOR task | task performance | Improved |
Closest papers by meaning
- Sign Gradient Descent-based Neuronal Dynamics: ANN-to-SNN Conversion Beyond ReLU Network (ICML 2024 · similarity 0.90)
- Learning Delays in Spiking Neural Networks using Dilated Convolutions with Learnable Spacings (ICLR 2024 · similarity 0.90)
- Take A Shortcut Back: Mitigating the Gradient Vanishing for Training Spiking Neural Networks (NeurIPS 2024 · similarity 0.89)
- Advancing Training Efficiency of Deep Spiking Neural Networks through Rate-based Backpropagation (NeurIPS 2024 · similarity 0.89)
- NDOT: Neuronal Dynamics-based Online Training for Spiking Neural Networks (ICML 2024 · similarity 0.89)
- Rethinking the Membrane Dynamics and Optimization Objectives of Spiking Neural Networks (NeurIPS 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.
