Research map / Label-efficient and robust learning / Recognition and long-tail
Transductive Bounds for the Multi-Class Majority Vote Classifier
Vasilii Feofanov, Emilie Devijver, Massih-Reza Amini
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 |
|---|---|---|---|
| learn pseudo-label margin thresholds from the transductive bound (MSLA) instead of fixed threshold 0.7 (FSLA) (ablation) | multi-class self-learning with Random Forest · MNIST | accuracy: .7998 -> .9085 | Improved |
| learn thresholds from the bound (MSLA) instead of fixed 0.7 (ablation) | multi-class self-learning · Pendigits | accuracy: .8474 -> .8866 | Improved |
| learn thresholds from the bound (MSLA) instead of fixed 0.7 (ablation) | multi-class self-learning · SensIT | accuracy: .6192 -> .6745 | Improved |
| learn thresholds from the bound (MSLA) instead of fixed 0.7 (ablation) | multi-class self-learning · Vowel | accuracy: .516 -> .5918 | Improved |
| add MSLA self-training on unlabeled data to supervised Random Forest | Random Forest · MNIST | accuracy: .8647 -> .9085 | Improved |
| add MSLA self-training to supervised Random Forest | Random Forest · Vowel, SensIT | accuracy: .5832 -> .5918 (Vowel); .67 -> .6745 (SensIT) | No clear effect |
| use MSLA instead of OVA-TSVM | semi-supervised classifier · DNA | accuracy: .8125 -> .7059 | Hurt |
| add fixed-threshold self-training FSLA to Random Forest | Random Forest · MNIST | accuracy: .8647 -> .7998 | Hurt |
Closest papers by meaning
- Multi-class Probabilistic Bounds for Majority Vote Classifiers with Partially Labeled Data (JMLR 2024 · similarity 0.95)
- Second Order PAC-Bayesian Bounds for the Weighted Majority Vote (NeurIPS 2020 · similarity 0.88)
- How many classifiers do we need? (NeurIPS 2024 · similarity 0.83)
- Multiclass Transductive Online Learning (NeurIPS 2024 · similarity 0.83)
- Statistical Query Hardness of Multiclass Linear Classification with Random Classification Noise (ICML 2025 · similarity 0.82)
- A Trichotomy for List Transductive Online Learning (ICML 2025 · similarity 0.82)
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.
