AI Paper Atlas

Research map / Architectures and efficiency / Neural architecture search

BN-NAS: Neural Architecture Search With Batch Normalization

Boyu Chen, Peixia Li, Baopu Li, Chen Lin, Chuming Li, Ming Sun, Junjie Yan, Wanli Ouyang

ICCV 2021 · Architectures and efficiency › Neural architecture search · approach: Efficient NAS search strategies · 35 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
use BN-indicator instead of validation accuracy to rank subnets and train only 10 epochsSPOS · ImageNet valtop-1 accuracy (%): 75.73 -> 75.67No clear effect
use BN-indicator and 15 epochs of BN-only supernet trainingFairNAS · ImageNet valtop-1 accuracy (%): 74.07 -> 74.12No clear effect
replace validation-accuracy subnet search by BN-indicatorSPOS · ImageNet search costsubnet search cost: ~1 GPU day -> 0.14s on CPUImproved
reduce supernet epochs and train only BN layersSPOS · ImageNet supernet trainingepochs: 100 -> 10Improved
train only BN parameters instead of all parameters (ablation)SPOS supernet · ImageNet supernet trainingtraining time: about 20% speedupImproved
use BN-indicator instead of accuracy indicator with 100 epochs of all-parameter training (ablation)SPOS · ImageNetsearched subnet accuracy: All/BN/100 vs All/Acc/100 similarNo clear effect
use BN-indicator instead of accuracy indicator with 30 epochs (ablation)SPOS · ImageNetsearched subnet accuracy: All/BN/30 better than All/Acc/30Improved
use accuracy indicator with 30 epoch supernet versus random baseline (ablation)SPOS · ImageNetsearched subnet accuracy: +0.1% over random baselineNo clear effect
BN-indicator with all-parameter training cut to 10 epochs (ablation)SPOS · ImageNetsearched subnet accuracy: 10 epochs All/BN fails to keep performanceHurt
train only BN parameters for 10 epochs (ablation)BN-NAS · ImageNetsearched subnet accuracy: BN/BN/10 recovers performanceImproved
BN-indicator versus accuracy on rank correlation with retrain accuracy (ablation)SPOS · CIFAR10Kendall Tau: similar Kendall TauNo clear effect
add SE modules to BN-NAS SPOS modelSPOS(Ours) · ImageNet valtop-1 accuracy (%): 75.67 -> 76.78Improved

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