AI Paper Atlas

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

MaxMatch: Semi-Supervised Learning With Worst-Case Consistency

Yangbangyan Jiang, Xiaodan Li, Yuefeng Chen, Yuan He, Qianqian Xu, Zhiyong Yang, Xiaochun Cao, Qingming Huang

TPAMI 2022 · Label-efficient and robust learning › Semi-supervised learning · approach: Safe and realistic semi-supervised learning · 36 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 worst-case consistency over K augmented variants to FixMatchFixMatch · ImageNet-1k (10% labels, from scratch)top-1 error (%): 28.74 -> 27.83Improved
add worst-case consistency to FixMatch with MoCo v2 pretrainingFixMatch · ImageNet-1k (MoCo v2 pretrain)top-1 error (%): 28.25 -> 28.09Improved
add worst-case consistency to FixMatch with MoCo v2 and EMANFixMatch+EMAN · ImageNet-1k (MoCo v2 + EMAN)top-1 error (%): 26.56 -> 26.09Improved
raise initial learning rate from 0.01 to 0.03 for FixMatch (ablation)FixMatch · ImageNet-1ktop-1 error (%): 29.70 -> 28.74Improved
raise initial learning rate from 0.02 to 0.03 for MaxMatch (ablation)MaxMatch · ImageNet-1ktop-1 error (%): 27.83 -> 28.48Hurt
MaxMatch advantage over FixMatch at lr 0.01 (ablation)MaxMatch · ImageNet-1k (lr 0.01)top-1 error (%): 29.70 -> 29.31Improved
MaxMatch advantage over FixMatch at lr 0.03 (ablation)MaxMatch · ImageNet-1k (lr 0.03)top-5 error (%): 10.94 -> 10.93No clear effect
use RandAugment instead of CTAugment in MaxMatch (ablation)MaxMatch · SVHN@40error rate (%): 1.92 -> 1.75Improved
use RandAugment instead of CTAugment in FixMatch (ablation)FixMatch · SVHN@40error rate (%): 7.65 -> 3.96Improved
add worst-case consistency to FixMatch (RandAugment) with 40 labels (ablation)FixMatch (RA) · SVHN@40error rate (%): 3.96 -> 1.75Improved
add worst-case consistency to FixMatch (CTAugment) with 40 labels (ablation)FixMatch (CTA) · SVHN@40error rate (%): 7.65 -> 1.92Improved
add worst-case consistency to FixMatch (RandAugment) with 1000 labels (ablation)FixMatch (RA) · SVHN@1000error rate (%): 2.28 -> 1.71Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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