Research map / Label-efficient and robust learning / Semi-supervised learning
Exploiting Synthetically Generated Data with Semi-Supervised Learning for Small and Imbalanced Datasets
M. Peréz-Ortiz, P. Tiňo, R. Mantiuk, C. Hervás-Martínez
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 synthetic convex-combination patterns as unlabelled data in S³VM | S³VM classification · Small high-dimensional datasets | accuracy: 53.00 -> 66.00 | Improved |
| Use class-dependent synthetic oversampling in S³VM instead of class-independent oversampling (ablation) | S³VM · Imbalanced UCI benchmark datasets, MCAR 0.8 | GM mean ranking: 3.65 -> 2.38 | Improved |
| Use an ensemble of S³VM models with different synthetic resampling distributions (ablation) | S³VM · Small-sample UCI benchmark datasets, MCAR 0.5 | MAcc mean ranking: 3.54 -> 2.69 | Improved |
| Vary the number of generated synthetic patterns (ablation) | S³VM-OvS · Synthetic data, d=100, N=50, V=0.5 | accuracy | No clear effect |
| Use real unlabelled data rather than synthetically generated unlabelled data (ablation) | S³VM · Synthetic datasets, especially high-dimensional settings | accuracy | Mixed |
| Apply class-independent instead of class-dependent synthetic oversampling (ablation) | S³VM · Imbalanced benchmark datasets | MAcc mean ranking: 2.69 -> 3.58; 2.67 -> 3.67 | Hurt |
| Use transductive graph-based methods to label synthetic data | Synthetic-data semi-supervised classification · Small-sample benchmark datasets | mean ranking: 8.67 -> 3.21 | Hurt |
Closest papers by meaning
- Data Augmentation with Diffusion for Open-Set Semi-Supervised Learning (NeurIPS 2024 · similarity 0.86)
- Maximizing the Potential of Synthetic Data: Insights from Random Matrix Theory (ICLR 2025 · similarity 0.86)
- Synthetic Tabular Data Generation for Imbalanced Classification: The Surprising Effectiveness of an Overlap Class (AAAI 2025 · similarity 0.85)
- Concentration and excess risk bounds for imbalanced classification with synthetic oversampling (NeurIPS 2025 · similarity 0.85)
- Scaling Up Semi-supervised Learning with Unconstrained Unlabelled Data (AAAI 2024 · similarity 0.84)
- SimPro: A Simple Probabilistic Framework Towards Realistic Long-Tailed Semi-Supervised Learning (ICML 2024 · similarity 0.84)
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.
