Research map / Label-efficient and robust learning / Semi-supervised learning
Robust Semi-Supervised Learning when Not All Classes have Labels
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 entropy prior regularizer L_ENT to BCE (ablation) | NACH basic model · CIFAR-100 | all-class accuracy: 31.5 -> 50.7 | Helped |
| add similarity filter to BCE (FBCE) with L_ENT, i.e. full L_UC (ablation) | NACH basic model · CIFAR-100 | all-class accuracy: 50.7 -> 50.2 | No clear effect |
| add L_UC loss to FixMatch-style SSL (ablation) | basic model · CIFAR-100 | all-class accuracy: 20.3 -> 50.2 | Helped |
| add distribution alignment to UC model (ablation) | UC Model · CIFAR-100 | all-class accuracy: 50.2 -> 50.3 | No clear effect |
| add adaptive threshold alone to UC model (ablation) | UC Model · CIFAR-100 | all-class accuracy: 50.2 -> 49.6 (all); 44.3 -> 45.6 (unseen) | Mixed |
| add adaptive threshold with distribution alignment (DTA) to UC model (ablation) | UC Model · CIFAR-100 | all-class accuracy: 50.2 -> 52.1 | Helped |
| NACH vs ORCA, unseen classes | ORCA · CIFAR-100 | unseen accuracy: 43.0 -> 47.0 | Helped |
| similarity filter (FBCE) vs plain BCE for pair selection (ablation) | basic model · CIFAR-100 | wrongly selected seen-unseen pairs | Helped |
| vary filter threshold k (ablation) | NACH · CIFAR-100 | accuracy | No clear effect |
| vary adaptive threshold margin beta (ablation) | NACH · CIFAR-100 | accuracy | No clear effect |
Closest papers by meaning
- Safe Deep Semi-Supervised Learning for Unseen-Class Unlabeled Data (ICML 2020 · similarity 0.92)
- Re-Evaluating the Impact of Unseen-Class Unlabeled Data on Semi-Supervised Learning Model (ICLR 2025 · similarity 0.91)
- Safe-Student for Safe Deep Semi-Supervised Learning With Unseen-Class Unlabeled Data (CVPR 2022 · similarity 0.91)
- On Non-Random Missing Labels in Semi-Supervised Learning (ICLR 2022 · similarity 0.90)
- OwMatch: Conditional Self-Labeling with Consistency for Open-World Semi-Supervised Learning (NeurIPS 2024 · similarity 0.89)
- InterLUDE: Interactions between Labeled and Unlabeled Data to Enhance Semi-Supervised Learning (ICML 2024 · similarity 0.87)
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.