AI Paper Atlas

Research map / Architectures and efficiency / Quantization

SiMaN: Sign-to-Magnitude Network Binarization

Mingbao Lin, Rongrong Ji, Zihan Xu, Baochang Zhang, Fei Chao, Chia‐Wen Lin, Ling Shao

TPAMI 2022 · Architectures and efficiency › Quantization · approach: Binary neural networks · 32 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 half-half binarization (top-half magnitude to +1) on top of l2-regularized weights (ablation)SiMaN · ResNet-18 ImageNettop-1 accuracy (%): 55.1 -> 59.2Improved
remove l2 regularization on binarized weights (no half-half) (ablation)SiMaN · ResNet-18 ImageNettop-1 accuracy (%): 55.1 -> 57.3Improved
remove l2 regularization with half-half binarization (ablation)SiMaN · ResNet-18 ImageNettop-1 accuracy (%): 59.2 -> 60.1Improved
remove l2 regularization with half-half binarization (ablation)SiMaN · ResNet-18 ImageNettop-5 accuracy (%): 81.5 -> 82.3Improved
add half-half binarization without l2 regularization (ablation)SiMaN · ResNet-18 ImageNettop-1 accuracy (%): 57.3 -> 60.1Improved
replace RBNN sign-based binarization with SiMaNbinarized ResNet-18 · ImageNettop-1 accuracy (%): 59.9 -> 60.1Improved
replace RBNN with SiMaNbinarized ResNet-34 · ImageNettop-1 accuracy (%): 63.1 -> 63.9Improved
replace RBNN with SiMaNbinarized VGG-small · CIFAR-10top-1 accuracy (%): 91.3 -> 92.5Improved
replace RBNN with SiMaNbinarized ResNet-18 · CIFAR-10top-1 accuracy (%): 92.2 -> 92.5Improved
replace RBNN with SiMaNbinarized ResNet-20 · CIFAR-10top-1 accuracy (%): 87.8 -> 87.4Hurt
use SiMaN on ReActNet structure instead of ReActNet trainingReActNet ResNet-18 · ImageNettop-1 accuracy (%): 65.9 -> 66.1Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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