Research map / Learning theory and algorithms / Prediction and decision losses
Popular decision tree algorithms are provably noise tolerant
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 |
|---|---|---|---|
| Train an impurity-based decision tree using greedy purity-gain splits | TopDownDT · Arbitrary distributions under nasty noise | allowable noise rate while achieving error at most ε: η ≤ O(εγ) | Helped |
| Use impurity-based decision trees at noise rates at or above the lower-bound regime | TopDownDT · Uniform features over {0,1}^d | tree size required to achieve error ≤ ε: η ≥ Ω(γε log(1/ε)); t ≥ 2^{d - O(log(1/γ)/γ)} | Hurt |
| Apply impurity-based decision trees to monotone size-s targets over product feature distributions | TopDownDT · Product distributions over {0,1}^d; monotone size-s decision trees | runtime under nasty noise: poly(d)·s^{O((log s)/ε^2)} | Helped |
Closest papers by meaning
- Robust Loss Functions for Training Decision Trees with Noisy Labels (AAAI 2024 · similarity 0.89)
- Optimal Decision Tree Pruning Revisited: Algorithms and Complexity (ICML 2025 · similarity 0.85)
- Near-Optimal Decision Trees in a SPLIT Second (ICML 2025 · similarity 0.85)
- Fully-Dynamic Approximate Decision Trees With Worst-Case Update Time Guarantees (ICML 2024 · similarity 0.85)
- Improving Decision Trees through the Lens of Parameterized Local Search (NeurIPS 2025 · similarity 0.84)
- Learning Small Decision Trees with Few Outliers: A Parameterized Perspective (AAAI 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.