Research map / Label-efficient and robust learning / Semi-supervised learning
Self-Paced Co-training
Fan Ma, Deyu Meng, Qi Xie, Zina Li, Xuanyi Dong
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 changed | Where | Measured | Effect |
|---|---|---|---|
| SPaCo vs standard Cotrain, 1 labeled step | Cotrain · Text classification average, k=1 | accuracy: 0.7987 -> 0.8151 | Improved |
| SPaCo vs Cotrade at k=3 | Cotrade · Text classification average, k=3 | accuracy: 0.8550 -> 0.8644 | Improved |
| SPaCo vs Cotrade at k=2 | Cotrade · Text classification average, k=2 | accuracy: 0.8348 -> 0.8413 | Improved |
| SPaCo vs RANC on ng1 k=1 | RANC · ng1 (k=1) | accuracy: 0.6170 -> 0.5970 | Hurt |
| SPaCo vs RANC on ad12 k=1 | RANC · ad12 (k=1) | accuracy: 0.9336 -> 0.9253 | Hurt |
| SPaCo vs SelfTrain average k=1 | SelfTrain · Text classification average, k=1 | accuracy: 0.7687 -> 0.8151 | Improved |
| increase labeled instances k=1 to 3 | SPaCo · Text classification average | accuracy: 0.8151 -> 0.8644 | Improved |
| SPaCo vs Cotrain on course k=1 | Cotrain · course (k=1) | accuracy: 0.8820 -> 0.9121 | Improved |
| SPaCo vs Cotrade on ng1 k=3 | Cotrade · ng1 (k=3) | accuracy: 0.7072 -> 0.6827 | Hurt |
| SPaCo vs base on Market-1501 Alexnet&Googlenet | Base · Market-1501 final | rank-1 accuracy: 0.5537 -> 0.6785 | Improved |
| SPaCo vs SelfTrain on Market-1501 Alexnet&Googlenet | SelfTrain · Market-1501 final | rank-1 accuracy: 0.6603 -> 0.6785 | Improved |
| SPaCo vs Cotrain on Market-1501 Vggnet&Googlenet | Cotrain · Market-1501 final | rank-1 accuracy: 0.6726 -> 0.6825 | Improved |
Closest papers by meaning
- Self-paced Multi-view Co-training (JMLR 2020 · similarity 0.91)
- Semi-supervised Learning with Multi-Head Co-Training (AAAI 2022 · similarity 0.86)
- Curriculum Labeling: Revisiting Pseudo-Labeling for Semi-Supervised Learning (AAAI 2021 · similarity 0.85)
- A Survey on Curriculum Learning (TPAMI 2021 · similarity 0.84)
- Fast Co-Training under Weak Dependence via Stream-Based Active Learning (ICML 2024 · similarity 0.83)
- Co-rewarding: Stable Self-supervised RL for Eliciting Reasoning in Large Language Models (ICLR 2026 · similarity 0.83)
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.
