AI Paper Atlas

Research map / Architectures and efficiency / Spiking networks

Efficiently Training Time-to-First-Spike Spiking Neural Networks from Scratch

Kaiwei Che, Zhengyu Ma, Yifan Huang, Peng Xue, Li Yuan, Wei Fang, Yonghong Tian

ICML 2026 · Architectures and efficiency › Spiking networks · approach: Training deep spiking networks · 0 citations (OpenAlex)

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 changedWhereMeasuredEffect
add ETTFS-init, average-pooling, weight normalization with affine and temporal weighting decoder together (ablation)SCNN5 TTFS SNN · Fashion-MNISTaccuracy (%): 89.61 -> 92.90Improved
replace Kaiming init with ETTFS-init (ablation)SCNN5 TTFS SNN · Fashion-MNISTaccuracy and convergence speedImproved
add weight normalization on top of ETTFS-init (ablation)SCNN5 TTFS SNN · Fashion-MNIST, CIFAR10, DVS GestureaccuracyImproved
use ETTFS-init instead of Kaiming init in deeper networks (5 to 11 layers) (ablation)TTFS SNN · Fashion-MNISTaccuracyImproved
use ETTFS-init instead of Kaiming init for gradient scale (ablation)SCNN5 · weight gradientsgradient magnitude: Kaiming 1e-7 to 1e-10 -> ETTFS-init larger than 1e-3Improved
replace TQ-TTFS decoder with temporal weighting decoder (ablation)TTFS SNN · CIFAR10average inference time-steps: 4.62 -> 2.93Improved
replace TQ-TTFS decoder with temporal weighting decoder (ablation)TTFS SNN · CIFAR10estimated energy E1: 20.59uJ -> 12.19uJImproved
replace TQ-TTFS decoder with temporal weighting decoderTTFS SNN · DVS Gestureaccuracy (%): 88.89 -> 95.83Improved
use average-pooling instead of max-pooling (ablation)TTFS SNN · one-spike constraintspike constraint preservationImproved
increase decay factor gamma in temporal weighting decoder (ablation)temporal weighting decoder · Fashion-MNIST, CIFAR10, DVS Gestureaccuracy: optimal gamma = 3 for Fashion-MNIST (exp) and CIFAR10 (linear)Mixed
choose exponential vs linear decay in the decoder (ablation)temporal weighting decoder · DVS GestureaccuracyNo clear effect
initialize from pretrained ReLU weightsETTFS · CIFAR10accuracy (%): 90.56 -> 93.89Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy