AI Paper Atlas

Research map / Architectures and efficiency / Quantization

Symmetry Regularization and Saturating Nonlinearity for Robust Quantization

Sein Park, Yeongsang Jang, Eunhyeok Park

ECCV 2022 · Architectures and efficiency › Quantization · approach: Mixed-precision quantization · 1 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 SymReg to the baseline (ablation)MobileNet-V2 PTQ ACIQ · CIFAR-100 2-bit weighttop-1 accuracy (%): 6.92 -> 25.01Improved
add SatNL to the baseline (ablation)MobileNet-V2 PTQ ACIQ · CIFAR-100 2-bit weighttop-1 accuracy (%): 6.92 -> 14.45Improved
add SatNL to SymReg (ablation)MobileNet-V2 PTQ ACIQ · CIFAR-100 2-bit weighttop-1 accuracy (%): 25.01 -> 33.68Improved
add SatNL to ASAM (ablation)MobileNet-V2 PTQ ACIQ · CIFAR-100 2-bit weighttop-1 accuracy (%): 11.80 -> 25.77Improved
add ASAM to SymReg+SatNL (All) (ablation)MobileNet-V2 PTQ ACIQ · CIFAR-100 3-bit weighttop-1 accuracy (%): 69.66 -> 72.17Improved
add ASAM to SymReg+SatNL (All) (ablation)MobileNet-V2 PTQ ACIQ · CIFAR-100 2-bit weighttop-1 accuracy (%): 33.68 -> 39.27Improved
add SymReg and SatNL to full-precision training (ablation)MobileNet-V2 · CIFAR-100 FPtop-1 accuracy (%): 74.70 -> 74.41Hurt
add ASAM to SymReg+SatNL (All) (ablation)MobileNet-V2 · CIFAR-100 FPtop-1 accuracy (%): 74.41 -> 75.33Improved
add ASAM to KURE (ablation)MobileNet-V2 PTQ ACIQ · CIFAR-100 3-bit weighttop-1 accuracy (%): 71.26 -> 72.48Improved
add All to KURE (ablation)MobileNet-V2 PTQ ACIQ · CIFAR-100 3-bit weighttop-1 accuracy (%): 72.48 -> 73.34Improved
train with SymReg+SatNL+ASAM (ablation)ResNet-18 PTQ QDrop · ImageNet W4/A4top-1 accuracy (%): 68.91 -> 69.93Improved
train with SymReg+SatNL+ASAMMobileNet-V2 PTQ ACIQ · ImageNet W4/A4top-1 accuracy (%): 18.15 -> 58.30Improved

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