AI Paper Atlas

Research map / Label-efficient and robust learning / Semi-supervised learning

MarginMatch: Improving Semi-Supervised Learning with Pseudo-Margins

Tiberiu Sosea, Cornelia Caragea

CVPR 2023 · Label-efficient and robust learning › Semi-supervised learning · approach: Confidence-based pseudo-label selection · 24 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 training-dynamics pseudo-margin filtering to flexible-threshold pseudo-labelingFlexMatch-style SSL · CIFAR-100, 4 labels/classerror rate: 39.99 -> 36.97Improved
add training-dynamics pseudo-margin filtering to flexible-threshold pseudo-labelingFlexMatch-style SSL · CIFAR-100, 25 labels/classerror rate: 26.96 -> 23.71Improved
add training-dynamics pseudo-margin filtering to flexible-threshold pseudo-labelingFlexMatch-style SSL · CIFAR-10, all reported label countserror rate: 5.04 -> 4.91; 5.04 -> 4.73; 4.19 -> 3.98Improved
add training-dynamics pseudo-margin filtering to flexible-threshold pseudo-labelingFlexMatch-style SSL · STL-10, 4 labels/classerror rate: 29.15 -> 25.37Improved
add training-dynamics pseudo-margin filtering to flexible-threshold pseudo-labelingFlexMatch-style SSL · SVHN, 25 labels/classerror rate: 3.13 -> 3.14No clear effect
use EMA smoothing δ=0.95 instead of δ=0.997 (ablation)MarginMatch · CIFAR-100, 4 labels/classerror rate: 37.91 -> 38.13Hurt
use EMA smoothing δ=0.99 instead of δ=0.997 (ablation)MarginMatch · CIFAR-100, 4 labels/classerror rate: 37.91 -> 38.05Hurt
use EMA smoothing δ=0.995 instead of δ=0.997 (ablation)MarginMatch · CIFAR-100, 4 labels/classerror rate: 37.91 -> 37.92No clear effect
use EMA smoothing δ=0.999 instead of δ=0.997 (ablation)MarginMatch · CIFAR-100, 4 labels/classerror rate: 37.91 -> 39.12Hurt
replace EMA with simple pseudo-margin averaging (δ=1) (ablation)MarginMatch · CIFAR-100, 4 labels/classerror rate: 37.91 -> 39.72Hurt
replace pseudo-margin with average confidence (ablation)MarginMatch · CIFAR-100, 4 labels/classerror rate: 36.97 -> 41.23Hurt
replace pseudo-margin with average entropy (ablation)MarginMatch · CIFAR-100, 4 labels/classerror rate: 36.97 -> 45.10Hurt

Later papers that compared with this method

1 later accepted paper reports MarginMatch as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 2 on most shared results.

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