AI Paper Atlas

Research map / Label-efficient and robust learning / Self-supervised and contrastive

Optimal Representations for Covariate Shift

Yangjun Ruan, Yann Dubois, Chris J. Maddison

ICLR 2022 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Invariance-based self-supervised learning · 11 citations (OpenAlex)

Read the paper

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
Fine-tune CLIP L with CAD (ablation)CLIP L · DomainBed benchmarksaccuracy: 80.7 $\pm$ 0.4 -> 81.4 $\pm$ 0.8 (VLCS); 93.7 $\pm$ 0.8 -> 94.7 $\pm$ 0.4 (PACS); 79.9 $\pm$ 0.1 -> 80.2 $\pm$ 0.2 (OffiMixed
Fine-tune CLIP L on LAION with an entropy bottleneck instead of without one (ablation)CLIP L · ImageNet distribution-shift datasetsaccuracy: 73.8 -> 74.2 (IN); 62.1 -> 62.7 (IN-V2); 37.0 -> 38.9 (IN-S); 56.9 -> 58.1 (YT-BB); 68.8 -> 70.1 (IN-Vid); 41.3 -> 42.1 Improved
Fine-tune CLIP L on LAION without an entropy bottleneck (ablation)CLIP L · ImageNet distribution-shift datasetsaccuracy: 75.2 -> 73.8 (IN); 64.2 -> 62.1 (IN-V2); 41.0 -> 37.0 (IN-S); 58.4 -> 56.9 (YT-BB); 71.6 -> 68.8 (IN-Vid); 42.8 -> 41.3 Hurt
Use domain-agnostic augmentations rather than non-domain-agnostic standard or intra-domain augmentations (ablation)CAD representation learning · PACSworst-case log likelihoodImproved
Use approximately domain-agnostic augmentations rather than strictly domain-agnostic augmentations (ablation)CAD representation learning · PACSworst-case log likelihoodNo clear effect
Increase the domain bottleneck weight (ablation)CAD/Ent representation learning · PACSworst-case target log likelihoodImproved
Increase the domain bottleneck weight beyond approximately 10^2 (ablation)CAD/Ent representation learning · PACSsource performanceHurt
Add a domain bottleneck (ablation)Representation learning on PACS · PACSworst-case target log likelihoodImproved
Train representations without access to one target domain (ablation)CAD representation learning · PACSworst-case log likelihood: -0.8 $\pm$ 0.2 -> -4.2 $\pm$ 0.2Hurt

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