Research map / Label-efficient and robust learning / Semi-supervised learning
Boosting Semi-Supervised Learning by Exploiting All Unlabeled Data
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 changed | Where | Measured | Effect |
|---|---|---|---|
| Add EML and ANL to FixMatch (ablation) | FixMatch · CIFAR-100, 400 labels | accuracy: 57.68 -> 59.32 (+1.64) | Helped |
| Add EML implemented with CE (ablation) | FixMatch · CIFAR-100, 400 labels | accuracy: 57.68 -> 58.35 (+0.67) | Helped |
| Add EML implemented with BCE (ablation) | FixMatch · CIFAR-100, 400 labels | accuracy: 57.68 -> 58.47 (+0.79) | Helped |
| Apply ANL only to examples with pseudo-labels (ablation) | FixMatch · CIFAR-100, 400 labels | accuracy: 57.68 -> 57.83 (+0.15) | Helped |
| Apply ANL only to examples without pseudo-labels (ablation) | FixMatch · CIFAR-100, 400 labels | accuracy: 57.68 -> 58.59 (+0.91) | Helped |
| Apply ANL to examples with and without pseudo-labels (ablation) | FixMatch · CIFAR-100, 400 labels | accuracy: 57.68 -> 58.67 (+0.99) | Helped |
| Vary α and β loss weights from 0.5 to 2.0 (ablation) | FullMatch · CIFAR-100, 10000 labels | accuracy: 78.31 to 78.50 | No clear effect |
| Add EML to FixMatch | FixMatch · CIFAR-100 test set | low-entropy prediction proportion: 60% -> 78% | Helped |
Closest papers by meaning
- FixMatch: Simplifying Semi-Supervised Learning with Consistency and Confidence (NeurIPS 2020 · similarity 0.93)
- AlphaMatch: Improving Consistency for Semi-Supervised Learning With Alpha-Divergence (CVPR 2021 · similarity 0.92)
- Barely-Supervised Learning: Semi-supervised Learning with Very Few Labeled Images (AAAI 2022 · similarity 0.92)
- CGMatch: A Different Perspective of Semi-supervised Learning (CVPR 2025 · similarity 0.91)
- Bi-Level Optimization for Semi-Supervised Learning with Pseudo-Labeling (AAAI 2025 · similarity 0.90)
- InterLUDE: Interactions between Labeled and Unlabeled Data to Enhance Semi-Supervised Learning (ICML 2024 · similarity 0.90)
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.