Research map / Learning theory and algorithms / Prediction and decision losses
Online Agnostic Multiclass Boosting
Vinod Raman, Ambuj Tewari
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 online agnostic boosting (Agn) instead of OnlineMBBM | OnlineMBBM · UCI Balance | accuracy: 79.5 -> 84.4 | Improved |
| use Agn instead of OnlineMBBM | OnlineMBBM · UCI Cars | accuracy: 69.9 -> 80.6 | Improved |
| use Agn instead of OnlineMBBM | OnlineMBBM · UCI Mice | accuracy: 77.6 -> 86.0 | Improved |
| use Agn instead of OnlineMBBM | OnlineMBBM · UCI Landsat | accuracy: 80.8 -> 67.0 | Hurt |
| use Agn instead of OnlineMBBM | OnlineMBBM · UCI Abalone | accuracy: 24.9 -> 22.1 | Hurt |
| use Agn instead of OnlineMBBM | OnlineMBBM · UCI Abalone | runtime (s): 9775.3 -> 4036.7 | Improved |
| use OCO-based realizable boosting (OCOR) instead of OnlineMBBM | OnlineMBBM · UCI Segment | accuracy: 78.9 -> 79.3 | No clear effect |
| use OCOR instead of OnlineMBBM | OnlineMBBM · UCI Landsat | runtime (s): 1814.3 -> 440.4 | Improved |
| use OCOR instead of OnlineMBBM | OnlineMBBM · UCI Yeast | accuracy: 39.8 -> 47.6 | Improved |
| use Adaboost.OLM instead of OnlineMBBM | OnlineMBBM · UCI Landsat | runtime (s): 1814.3 -> 81.8 | Improved |
| use Adaboost.OLM instead of OnlineMBBM | OnlineMBBM · UCI Landsat | accuracy: 80.8 -> 56.9 | Hurt |
| use random relabeling instead of fractional relabeling (ablation) | Agn · UCI datasets | accuracy and speed | Mixed |
Closest papers by meaning
- Online Agnostic Boosting via Regret Minimization (NeurIPS 2020 · similarity 0.91)
- Online multiclass boosting (NeurIPS 2017 · similarity 0.90)
- Revisiting Agnostic Boosting (NeurIPS 2025 · similarity 0.89)
- Sample-Optimal Agnostic Boosting with Unlabeled Data (ICML 2025 · similarity 0.84)
- A Boosting-Type Convergence Result for AdaBoost.MH with Factorized Multi-Class Classifiers (NeurIPS 2024 · similarity 0.83)
- The Many Faces of Optimal Weak-to-Strong Learning (NeurIPS 2024 · 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.
