AI Paper Atlas

Research map / Architectures and efficiency / Neural architecture search

Fair DARTS: Eliminating Unfair Advantages in Differentiable Architecture Search

Xiangxiang Chu, Tianbao Zhou, Bo Zhang, Jixiang Li

ECCV 2020 · Architectures and efficiency › Neural architecture search · approach: Bi-level differentiable NAS · 357 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 zero-one loss (ablation)FairDARTS · CIFAR-10 test, S1top-1 accuracy (%): 97.33 -> 97.41Improved
add zero-one loss (ablation)FairDARTS · CIFAR-10 test, S1multiply-adds (M): 532 -> 458Improved
add zero-one loss to softmax DARTS (ablation)DARTS · CIFAR-10 test, S1top-1 accuracy (%): 96.76 -> 96.77No clear effect
replace softmax with independent sigmoid (collaborative competition) (ablation)DARTS · ImageNet S2 searchtop-1 accuracy of derived model (%): 66.4 -> 75.6Improved
remove skip connections from search space (ablation)DARTS · CIFAR-10 test, S1top-1 accuracy (%): 96.76 -> 96.88Improved
use FairDARTS instead of DARTS without skip connectionsS1 · CIFAR-10 testtop-1 accuracy (%): 96.88 -> 97.41Improved
add Gaussian noise to skip-connection architecture weights (ablation)DARTS · CIFAR-10 test, S1top-1 accuracy (%): 97.12Improved
restrict skip connections to M=2 and sample randomly (ablation)random search · CIFAR-10 test, S1top-1 accuracy (%): 97.01 -> 97.14Improved
use FairDARTS-C with SE, Swish and AutoAugment instead of EfficientNet B0FairDARTS-C · ImageNet valtop-1 accuracy (%): 76.3 -> 77.2Improved
use FairDARTS-D instead of PC-DARTSImageNet S2 · ImageNet valtop-1 accuracy (%): 74.9 -> 75.6Improved
use FairDARTS-C backbone instead of MixNet-MRetinaNet · COCO val2017AP (%): 31.3 -> 31.9Improved
raise sigma threshold from 0.75 to 0.8 (ablation)FairDARTS · ImageNet valparams / accuracy: 75.6 -> 75.1Mixed

Later papers that compared with this method

7 later accepted papers report FairDARTS as a baseline in their results tables; 7 of them report a better number on most of the results they share with it.

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