Research map / Learning theory and algorithms / Sample complexity
AdaBoost is not an Optimal Weak to Strong Learner
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Run AdaBoost (and variants AdaBoost_nu, AdaBoost*_nu, DualLPboost) with an adversarial weak learner that knows the unsampled points and accumulates errors outside the training set | AdaBoost and classic boosting variants that invoke the weak learner on the full training set · theoretical sample complexity (PAC weak-to-strong learning) | sample complexity m(eps) for error eps | Hurt |
| Weak learner receives distributions with support on only a random constant fraction of training data (subsampling) instead of full training set | boosting algorithm invoking weak learner · theoretical sample complexity argument | sample complexity lower bound applicability | No clear effect |
Closest papers by meaning
- The Many Faces of Optimal Weak-to-Strong Learning (NeurIPS 2024 · similarity 0.92)
- A Boosting-Type Convergence Result for AdaBoost.MH with Factorized Multi-Class Classifiers (NeurIPS 2024 · similarity 0.87)
- Revisiting Agnostic Boosting (NeurIPS 2025 · similarity 0.86)
- Robust Minimax Boosting with Performance Guarantees (NeurIPS 2025 · similarity 0.85)
- Optimal Parallelization of Boosting (NeurIPS 2024 · similarity 0.84)
- Provable weak-to-strong generalization via benign overfitting (ICLR 2025 · 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.