AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

EagleEye: Fast Sub-net Evaluation for Efficient Neural Network Pruning

Bailin Li, Bowen Wu, Jiang Su, Guangrun Wang

ECCV 2020 · Architectures and efficiency › Pruning and efficient tuning · approach: Structured network pruning

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 adaptive BN statistics recalibration before evaluating pruned candidates (ablation)MobileNetV1 L1 pruning · CIFAR-10, FLOPs not fixedPearson correlation: 0.613 -> 0.933Improved
add adaptive BN recalibration (ablation)MobileNetV1 L1 pruning · CIFAR-10, 50% FLOPsPearson correlation: 0.638 -> 0.854Improved
add adaptive BN recalibration (ablation)MobileNetV1 L1 pruning · CIFAR-10, 62.5% FLOPsPearson correlation: 0.655 -> 0.857Improved
add adaptive BN recalibration (ablation)ResNet-50 L1 pruning · CIFAR-10Pearson correlation: 0.279 -> 0.681Improved
add adaptive BN recalibration (ablation)MobileNetV2 L1 pruning · ImageNetPearson correlation: -0.09 -> 0.634Improved
add adaptive BN recalibration (ablation)MobileNetV2 L1 pruning · ImageNetKendall tau correlation: -0.162 -> 0.404Improved
add adaptive BN recalibration (ablation)MobileNetV1 L1 pruning · CIFAR-10, FLOPs not fixedSpearman correlation: 0.689 -> 0.937Improved
add adaptive BN recalibration (ablation)MobileNetV1 L2-norm pruning · ImageNetPearson correlation: 0.074 -> 0.535Improved
add adaptive BN recalibration (ablation)MobileNetV1 L2-norm pruning · ImageNetKendall tau correlation: 0.014 -> 0.367Improved
add adaptive BN recalibration (ablation)ResNet-50 L1 pruning · CIFAR-10Spearman correlation: 0.450 -> 0.519Improved

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