Research map / Learning theory and algorithms / Prediction and decision losses
Optimal Decision Trees for Nonlinear Metrics
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 |
|---|---|---|---|
| Optimize the tree directly for F1-score instead of accuracy | Decision trees · seismic_bumps test, depth 3 | F1-score: 0.04 -> 0.36 | Helped |
| Optimize the tree directly for F1-score instead of accuracy | Decision trees · bank test, depth 3 | F1-score: 0.52 -> 0.56 | Helped |
| Optimize the tree directly for F1-score instead of accuracy | Decision trees · IndiansDiabetes test, depth 3 | F1-score: 0.70 -> 0.73 | Helped |
| Optimize the tree directly for F1-score instead of accuracy | Decision trees · banknote test, depth 3 | F1-score: 0.92 -> 0.92 | No clear effect |
| Add upper bounding (ablation) | Bi-objective decision-tree algorithm · diabetes, depth 4 | runtime (sec): 118 -> 73 | Helped |
| Add infeasibility-based lower bounds (ablation) | Bi-objective decision-tree algorithm · diabetes, depth 4 | runtime (sec): 109 -> 73 | Helped |
| Add similarity-based lower bounding (ablation) | Bi-objective decision-tree algorithm · diabetes, depth 4 | runtime (sec): 93 -> 73 | Helped |
| Add upper bounding (ablation) | Bi-objective decision-tree algorithm · anneal, depth 4 | runtime (sec): 20 -> 1397 | Hurt |
| Add infeasibility-based lower bounds (ablation) | Bi-objective decision-tree algorithm · anneal, depth 4 | runtime (sec): 18 -> 1397 | Hurt |
| Add similarity-based lower bounding (ablation) | Bi-objective decision-tree algorithm · anneal, depth 4 | runtime (sec): 11 -> 1397 | Hurt |
Closest papers by meaning
- Generalized and Scalable Optimal Sparse Decision Trees (ICML 2020 · similarity 0.89)
- Scalable Optimal Multiway-Split Decision Trees with Constraints (AAAI 2023 · similarity 0.87)
- Principled Algorithms for Optimizing Generalized Metrics in Binary Classification (ICML 2025 · similarity 0.85)
- Piecewise Constant and Linear Regression Trees: An Optimal Dynamic Programming Approach (ICML 2024 · similarity 0.85)
- Near-Optimal Decision Trees in a SPLIT Second (ICML 2025 · similarity 0.85)
- Learning Small Decision Trees with Few Outliers: A Parameterized Perspective (AAAI 2024 · similarity 0.85)
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.