AI Paper Atlas

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

Shrinking Class Space for Enhanced Certainty in Semi-Supervised Learning

Lihe Yang, Zhen Zhao, Lei Qi, Yu Qiao, Yinghuan Shi, Hengshuang Zhao

ICCV 2023 · Label-efficient and robust learning › Semi-supervised learning · approach: Confidence-based pseudo-label selection · 31 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
add ShrinkMatch uncertain-sample loss to FixMatch+DA baseline (ablation)FixMatch+DA · CIFAR-100 (400 labels)accuracy (%): 62.46 -> 64.64Improved
add ShrinkMatch to baseline (ablation)FixMatch+DA · STL-10 (40 labels)accuracy (%): 84.85 -> 85.98Improved
add both reweighting principles to shrunk loss (ablation)ShrinkMatch · CIFAR-100 (400 labels)accuracy (%): 63.16 -> 64.24Improved
replace adaptive reweighting with linear scheduling of loss weight (ablation)ShrinkMatch · CIFAR-100 (400 labels)accuracy (%): 64.24 -> 61.23Hurt
use auxiliary head instead of main head for shrunk-space loss (ablation)ShrinkMatch · CIFAR-100 (400 labels)accuracy (%): 60.65 -> 64.24Improved
shrink class space instead of using original space with aux head (ablation)ShrinkMatch · CIFAR-100 (400 labels)accuracy (%): 61.69 -> 64.24Improved
use hard label rather than soft label in shrunk space (ablation)ShrinkMatch · CIFAR-100 (400 labels)accuracy (%): 61.34 -> 64.24Improved
learn uncertain samples directly in original space with main head (ablation)FixMatch+DA baseline · CIFAR-100 (400 labels)accuracy (%): 60.96 -> 52.39Hurt
raise confidence threshold from 0.95 to 0.98 (ablation)ShrinkMatch · CIFAR-100 (400 labels)accuracy (%): 65.00 -> 65.80Improved
lower confidence threshold from 0.95 to 0.70 (ablation)ShrinkMatch · CIFAR-100 (400 labels)accuracy (%): 65.00 -> 59.58Hurt
ShrinkMatch vs SimMatch with 2500 labelsShrinkMatch · CIFAR-100 (2500 labels)accuracy (%): 74.93 -> 74.83No clear effect
ShrinkMatch vs FlexMatch with 250 labelsShrinkMatch · STL-10 (250 labels)accuracy (%): 91.77 -> 91.55No clear effect

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