AI Paper Atlas

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

Leveraging unlabeled data to predict out-of-distribution performance

Saurabh Garg, Sivaraman Balakrishnan, Zachary Chase Lipton, Behnam Neyshabur, Hanie Sedghi

ICLR 2022 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Mixture and distillation for domain shift · 19 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 ATC with a source-validation confidence threshold on unlabeled target data instead of the strongest prior baselineFixed classifier accuracy estimation · CIFAR10, CIFAR100, ImageNet, WILDS, MNIST, and BREEDS shiftsMAE: ATC typically achieves the smallest MAE; improves by more than 2× on CIFAR and 3–4× on ImageNet over GDEImproved
Apply source post-hoc Temperature Scaling before estimating accuracy (ablation)ATC and comparison methods · Multiple distribution shiftsMAEImproved
Use maximum confidence instead of negative entropy as the ATC score (ablation)ATC · CIFAR100 SyntheticMAE: 4.94 -> 4.75Improved
Use negative entropy instead of maximum confidence as the ATC score (ablation)ATC · ImageNet NaturalMAE: 2.17 -> 0.80Improved
Fit a robust linear regression to DOC using labeled same-subpopulation target shifts (ablation)DOC on BREEDS novel subpopulations · Living-17 novel subpopulationsMAE: 24.41 -> 13.26Improved
Use ATC without regression rather than fine-tuned DOCAccuracy estimation on BREEDS novel subpopulations · Living-17 novel subpopulationsMAE: 13.26 -> 10.22Improved
Shift the target invariant-feature class-conditional support while retaining a uniform distribution (ablation)ATC toy model · Toy model with shifted class-conditional supportaccuracy estimation biasHurt
Introduce novel rather than same subpopulationsAll accuracy-estimation methods · BREEDSMAEHurt
Pre-train the model before accuracy estimation (ablation)ATC · CIFARMAENo clear effect

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