AI Paper Atlas

Research map / Architectures and efficiency / Spiking networks

Error-Aware Conversion from ANN to SNN via Post-training Parameter Calibration

Yuhang Li, Shikuang Deng, Xin Dong, Shi Gu

IJCV 2024 · Architectures and efficiency › Spiking networks · approach: Training deep spiking networks · 46 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 ClipRound (bias offset) instead of floor (ablation)MaxAct SNN, VGG-16 no BN, T=32 · ImageNettop-1 accuracy: 0.1 -> 63.22Improved
add ClipRound instead of floor (ablation)MMSE SNN, ResNet-34 no BN, T=16 · ImageNettop-1 accuracy: 5.54 -> 28.70Improved
use MMSE threshold instead of percentile (ablation)SNN ResNet-34 no BN, T=16 · ImageNettop-1 accuracy: 15.09 -> 28.70Improved
use MMSE threshold instead of maximum activation (ablation)SNN ResNet-34 no BN, T=16 · ImageNettop-1 accuracy: 7.46 -> 28.70Improved
use channel-wise MMSE threshold instead of layer-wise (ablation)SNN ResNet-34 no BN, T=16 · ImageNettop-1 accuracy: 28.70 -> 43.69Improved
add light calibration pipeline (bias calibration) (ablation)baseline SNN VGG-16 with BN, T=16 · ImageNettop-1 accuracy: 0.95 -> 40.81Improved
add potential and weight calibration (ACP) on top of LCP (ablation)SNN VGG-16 with BN, T=16 · ImageNettop-1 accuracy: 40.81 -> 65.02Improved
combine ACP with maximum activation threshold instead of MMSE (ablation)SNN ResNet-34 with BN, T=16 · ImageNettop-1 accuracy: 50.84 -> 0.41Hurt
use ANN with BN layers instead of without BN, no calibration (ablation)MMSE + ClipRound SNN ResNet-34, T=16 · ImageNettop-1 accuracy: 28.70 -> 0.10Hurt
add ACP calibration (ablation)SNN MobileNet, T=128 · ImageNettop-1 accuracy: 1.99 -> 64.58Improved
add ACP calibration (ablation)SNN RetinaNet, T=32 · COCOmAP: 6.2 -> 28.4Improved
increase LCP calibration samples (ablation)LCP ResNet-34 · ImageNettop-1 accuracy: 57.23 -> 66.23Improved

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