AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Transferable Query Selection for Active Domain Adaptation

Bo Fu, Zhangjie Cao, Jianmin Wang, Mingsheng Long

CVPR 2021 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Feature alignment for unsupervised adaptation · 73 citations (OpenAlex)

Check 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 TQS instead of ADMA for selecting target samples with a 5% labeling budgetactive domain adaptation · Office-31accuracy (%): 89.4±0.3 -> 91.1±0.3Improved
use TQS instead of ADMA for selecting target samples with a 5% labeling budgetactive domain adaptation · Office-Homeaccuracy (%): 68.6±0.3 -> 70.5±0.3Improved
use TQS instead of ADMA for selecting target samples with a 5% labeling budgetactive domain adaptation · VisDA-2017accuracy (%): 81.4±0.4 -> 83.1±0.4Improved
add transferable committee to TQS (ablation)TQS · Office-31accuracyImproved
add transferable uncertainty to TQS (ablation)TQS · Office-31accuracyImproved
add transferable domainness to TQS (ablation)TQS · Office-31accuracyImproved
use min-max adversarial optimization for the committee (ablation)TQS · Office-31accuracyImproved
use transferable uncertainty rather than single-classifier entropy (ablation)TQS · Office-31accuracyImproved
use transferable domainness rather than domain similarity (ablation)TQS · Office-31accuracyImproved
use an ensemble of classifiers for uncertainty estimation (ablation)TQS · Office-31accuracyImproved
use margin rather than entropy as the uncertainty function (ablation)TQS · Office-31accuracyImproved
use randomized selection to diversify selected target samples (ablation)TQS · Office-31accuracyImproved

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