AI Paper Atlas

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

Scarf: Self-Supervised Contrastive Learning using Random Feature Corruption

Dara Bahri, Heinrich Jiang, Yi Tay, Donald Metzler

ICLR 2022 · Label-efficient and robust learning › Self-supervised and contrastive · approach: Contrastive versus non-contrastive theory · 10 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 SCARF contrastive pre-trainingControl neural network · OpenML-CC18, 100% labeledrelative gain in accuracy: 2.352Improved
Add SCARF contrastive pre-trainingControl neural network · OpenML-CC18, 25% labeledrelative gain in accuracy: 3.692Improved
Add SCARF contrastive pre-trainingControl neural network · OpenML-CC18, 30% label noiserelative gain in accuracy: 2.261Improved
Add SCARF pre-trainingDeep k-NN · OpenML-CC18, 30% label noiserelative gain in accuracy: 2.001Improved
Replace empirical marginal feature sampling with no corruption (ablation)SCARF pre-training · OpenML-CC18 classificationaccuracyHurt
Replace empirical marginal sampling with mean replacement (ablation)SCARF pre-training · OpenML-CC18 classificationaccuracyHurt
Replace marginal sampling with additive Gaussian noise (ablation)SCARF pre-training · OpenML-CC18 classificationaccuracyHurt
Replace independent marginal draws with joint training-data sampling (ablation)SCARF pre-training · OpenML-CC18 classificationaccuracyHurt
Replace marginal sampling with missing-feature values (ablation)SCARF pre-training · OpenML-CC18 classificationaccuracyHurt
Replace marginal sampling with feature dropout (ablation)SCARF pre-training · OpenML-CC18 classificationaccuracyHurt
Corrupt both views instead of only one (ablation)SCARF pre-training · OpenML-CC18 classificationaccuracyHurt
Use the same corrupted feature indices for every example in a batch (ablation)SCARF pre-training · OpenML-CC18 classificationaccuracyHurt

Later papers that compared with this method

1 later accepted paper reports SCARF as a baseline in their results tables; 1 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