AI Paper Atlas

Research map / Architectures and efficiency / Spiking networks

Fast-SNN: Fast Spiking Neural Network by Converting Quantized ANN

Yangfan Hu, Qian Zheng, Xudong Jiang, Gang Pan

TPAMI 2023 · Architectures and efficiency › Spiking networks · approach: Training deep spiking networks · 132 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 signed IF neuron model (ablation)native converted SNN (VGG-11 W2A2, T=3) · CIFAR-10top-1 accuracy: 84.92 -> 92.80Improved
add signed IF neuron model (ablation)native converted SNN (VGG-16 32/3, T=7) · ImageNettop-1 accuracy: 36.43 -> 72.89Improved
add layer-wise fine-tuning (ablation)SNN-beta (VGG-11 W2A2) · CIFAR-10top-1 accuracy: 92.80 -> 92.99Improved
add layer-wise fine-tuning (ablation)SNN-beta (VGG-16 32/3) · ImageNettop-1 accuracy: 72.89 -> 72.95Improved
add layer-wise fine-tuning (ablation)SNN-beta (2-bit ResNet-56) · CIFAR-10top-1 accuracy: 81.25 -> 89.25Improved
add layer-wise fine-tuning (ablation)SNN-beta (3-bit ResNet-56) · CIFAR-10top-1 accuracy: 88.55 -> 92.17Improved
add layer-wise fine-tuning (ablation)SNN-beta (3-bit ResNet-20) · CIFAR-10top-1 accuracy: 91.33 -> 92.18Improved
use learnable firing threshold instead of 99th percentile threshold (ablation)2-bit AlexNet SNN, T=3 · CIFAR-10top-1 accuracy: 81.66 -> 88.97Improved
use learnable firing threshold instead of 99th percentile threshold (ablation)2-bit VGG-11 SNN, T=3 · CIFAR-10top-1 accuracy: 80.48 -> 84.92Improved
use learnable firing threshold instead of max activation threshold (ablation)2-bit VGG-11 SNN, T=3 · CIFAR-10top-1 accuracy: 24.04 -> 84.92Improved
raise latency to 3x7 steps, removing sequential error, with learnable threshold (ablation)2-bit AlexNet SNN · CIFAR-10top-1 accuracy: 88.97 -> 91.52Improved
use 2-bit weights in Fast-SNN AlexNet instead of PTL's 4-bit weightsAlexNet · CIFAR-10top-1 accuracy: 89.48 -> 91.63Improved

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