Research map / Learning theory and algorithms / Prediction and decision losses
Hierarchical Shrinkage: Improving the accuracy and interpretability of tree-based models.
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 |
|---|---|---|---|
| Apply HS to RF | Random Forest · Diabetes | AUC: 0.733 -> 0.787 | Helped |
| Apply HS to CART | CART · Classification datasets, m = 15 | relative AUC: 6.2% | Helped |
| Apply HS to CART with CCP | CART with CCP · Classification datasets, m = 15 | relative AUC: 6.5% | Helped |
| Apply HS to C4.5 | C4.5 · Classification datasets, m = 15 | relative AUC: 8% | Helped |
| Apply HS to CART | CART · Regression datasets, m = 15 | relative R²: 9.8% | Helped |
| Apply HS to CART with CCP | CART with CCP · Regression datasets, m = 15 | relative R²: 10.1% | Helped |
| Replace leaf-based shrinkage with hierarchical shrinkage | CART · Real-world classification and regression datasets | predictive performance | Helped |
| Apply HS to RF rather than regularizing with tree depth or mtry | Random Forest · Real-world datasets | prediction accuracy | Helped |
| Apply HS to RF | Random Forest · RF ensemble-size comparison | number of trees needed for comparable performance: five times smaller | Helped |
| Use HS rather than BART | RF prediction · Classification and regression datasets | fit speed: 10-15 times faster | Helped |
| Apply HS to RF | Random Forest · Breast cancer and other classification datasets | SHAP-value stability | Helped |
Closest papers by meaning
- Multistage Defer Trees for Hybrid Interpretability: If at First You Can't Succeed, Tree Again (NeurIPS 2026 · similarity 0.89)
- Learning Small Decision Trees with Few Outliers: A Parameterized Perspective (AAAI 2024 · similarity 0.89)
- Invariant Random Forest: Tree-Based Model Solution for OOD Generalization (AAAI 2024 · similarity 0.88)
- Near-Optimal Decision Trees in a SPLIT Second (ICML 2025 · similarity 0.88)
- Optimal Decision Tree Pruning Revisited: Algorithms and Complexity (ICML 2025 · similarity 0.88)
- Symbolic Regression Enhanced Decision Trees for Classification Tasks (AAAI 2024 · similarity 0.88)
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.