AI Paper Atlas

Research map / Architectures and efficiency / CNN design

Lipschitz Continuity Retained Binary Neural Network

Yuzhang Shang, Dan Xu, Bin Duan, Ziliang Zong, Liqiang Nie, Yan Yan

ECCV 2022 · Architectures and efficiency › CNN design · approach: Theory of convolutional inductive bias · 24 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 LCR Lipschitz regularizer (ablation)IR-Net w/o BiReal · ImageNet ResNet-18top-1 accuracy (%): 56.9 -> 58.4Improved
add LCR regularizer (ablation)IR-Net with BiReal · ImageNet ResNet-18top-1 accuracy (%): 58.1 -> 59.6Improved
add LCR regularizer (ablation)ReActNet · ImageNettop-1 accuracy (%): 69.4 -> 69.8Improved
add LCR regularizer (ablation)IR-Net w/o BiReal · CIFAR-10 ResNet-20top-1 accuracy (%): 85.5 -> 86.0Improved
add LCR regularizer (ablation)IR-Net with BiReal · CIFAR-10 ResNet-20top-1 accuracy (%): 86.5 -> 87.2Improved
increase lambda from 0 to 2^3 (ablation)LCR-BNN ResNet-18 · CIFAR-10top-1 accuracy (%): 85.9 -> 91.8Improved
increase lambda from 0 to 2^2 (ablation)LCR-BNN ResNet-20 · CIFAR-10top-1 accuracy (%): 83.9 -> 87.2Improved
raise lambda from 2^2 to 2^3 (ablation)LCR-BNN ResNet-20 · CIFAR-10top-1 accuracy (%): 87.2 -> 86.5Hurt
add LCR regularizerIR-Net · ImageNet-CmCE (%): 89.2 -> 84.9Improved
add LCR regularizerRBNN · ImageNet-CmCE (%): 87.5 -> 84.8Improved
LCR instead of ProxyBNN (ResNet-18)LCR-BNN · ImageNettop-1 accuracy (%): 58.7 -> 59.6Improved
add LCR regularizer at inference costLCR-BNN · ResNet-18 BOPS/FLOPSBOPS and FLOPS: 1.676 x 10^9 -> 1.676 x 10^9No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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