Research map / Label-efficient and robust learning / Semi-supervised learning
Contrastive Pessimistic Likelihood Estimation for Semi-Supervised Classification
Marco Loog
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 |
|---|---|---|---|
| use MCPL semi-supervised LDA instead of supervised LDA | LDA · UCI landsat (test error) | test error rate: 0.291 -> 0.251 | Improved |
| use MCPL semi-supervised LDA instead of supervised LDA | LDA · UCI optical (test error) | test error rate: 0.161 -> 0.113 | Improved |
| use MCPL semi-supervised LDA instead of supervised LDA | LDA · UCI shuttle (test error) | test error rate: 0.134 -> 0.103 | Improved |
| use MCPL semi-supervised LDA instead of supervised LDA | LDA · UCI pen-based (test error) | test error rate: 0.280 -> 0.243 | Improved |
| use MCPL semi-supervised LDA instead of supervised LDA | LDA · UCI banknote (test error) | test error rate: 0.061 -> 0.052 | Improved |
| use MCPL semi-supervised LDA instead of supervised LDA | LDA · UCI letter (test error) | test error rate: 0.618 -> 0.599 | Improved |
| use MCPL semi-supervised LDA instead of supervised LDA | LDA · UCI spectf (test error) | test error rate: 0.325 -> 0.325 | No clear effect |
| use MCPL semi-supervised LDA instead of supervised LDA | LDA · UCI low (test error) | test error rate: 0.475 -> 0.501 | Hurt |
| use MCPL semi-supervised LDA instead of supervised LDA | LDA · UCI datasets (test log-likelihood) | log-likelihood: -1.88e+03 -> -62.6 | Improved |
| use MCPL semi-supervised LDA instead of supervised LDA | LDA · UCI spectf (training error) | training error rate: 0.203 -> 0.210 | Hurt |
| use MCPL semi-supervised LDA instead of ad hoc constrained LDA | constrained LDA · UCI optical (test error) | test error rate: 0.900 -> 0.113 | Improved |
| use ad hoc constrained LDA instead of MCPL | MCPL · UCI climate (test error) | test error rate: 0.143 -> 0.117 | Improved |
Closest papers by meaning
- SimPro: A Simple Probabilistic Framework Towards Realistic Long-Tailed Semi-Supervised Learning (ICML 2024 · similarity 0.87)
- Continuous Contrastive Learning for Long-Tailed Semi-Supervised Recognition (NeurIPS 2024 · similarity 0.86)
- Semi-Supervised Sparse Gaussian Classification: Provable Benefits of Unlabeled Data (NeurIPS 2024 · similarity 0.86)
- CGMatch: A Different Perspective of Semi-supervised Learning (CVPR 2025 · similarity 0.85)
- Hypothesis Testing for Class-Conditional Noise Using Local Maximum Likelihood (AAAI 2024 · similarity 0.85)
- Out-Of-Domain Unlabeled Data Improves Generalization (ICLR 2024 · similarity 0.85)
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.
