Research map / Bandits and online learning / Online convex optimization and games
Optimal Margin Distribution Learning in Dynamic Environments
Teng Zhang, Peng Zhao, Hai Jin
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 |
|---|---|---|---|
| D-ODM vs DWM | D-ODM · SEA | accuracy (%): 87.04 -> 87.49 | Improved |
| D-ODM vs SVM-win | D-ODM · SEA | accuracy (%): 73.94 -> 87.49 | Improved |
| D-ODM vs AdaBoost.OL | D-ODM · hyperplane | accuracy (%): 72.73 -> 89.29 | Improved |
| D-ODM vs SVM-ada | D-ODM · Weather | accuracy (%): 77.26 -> 79.11 | Improved |
| D-ODM vs DWM | D-ODM · UG-2C-5D | accuracy (%): 92.82 -> 92.11 | Hurt |
| D-ODM vs DWM | D-ODM · Electricity | accuracy (%): 78.60 -> 76.80 | Hurt |
| D-ODM vs SVM-win | D-ODM · MG-2C-2D | accuracy (%): 90.20 -> 90.16 | No clear effect |
| D-ODM vs AdaBoost.OL | D-ODM · Chess | accuracy (%): 78.39 -> 78.44 | No clear effect |
| D-ODM vs SVM-win | D-ODM · Usenet-1 | accuracy (%): 68.92 -> 68.98 | No clear effect |
| D-ODM vs DWM | D-ODM · UG-2C-2D | accuracy (%): 95.59 -> 95.99 | No clear effect |
| D-ODM vs SVM-fix | D-ODM · Electricity | accuracy (%): 62.01 -> 76.80 | Improved |
| D-ODM win/tie/loss against SVM-ada | D-ODM · 20 datasets | accuracy (%): 20/0/0 | Improved |
Closest papers by meaning
- Efficient Non-stationary Online Learning by Wavelets with Applications to Online Distribution Shift Adaptation (ICML 2024 · similarity 0.86)
- Dynamic Regret Reduces to Kernelized Static Regret (NeurIPS 2025 · similarity 0.85)
- Stochastic Bayesian Optimization with Unknown Continuous Context Distribution via Kernel Density Estimation (AAAI 2024 · similarity 0.85)
- An Equivalence Between Static and Dynamic Regret Minimization (NeurIPS 2024 · similarity 0.84)
- Non-stationary Online Learning for Curved Losses: Improved Dynamic Regret via Mixability (ICML 2025 · similarity 0.84)
- Sequential Probability Assignment with Contexts: Minimax Regret, Contextual Shtarkov Sums, and Contextual Normalized Maximum Likelihood (NeurIPS 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.
