Research map / Learning theory and algorithms / Prediction and decision losses
Pushing the Envelope of Gradient Boosting Forests via Globally-Optimized Oblique Trees
Check 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 |
|---|---|---|---|
| Use globally optimized oblique trees instead of axis-aligned trees | Gradient boosting forests · MNIST | test error (%): 1.96±0.03 -> 1.55±0.02 | Helped |
| Use globally optimized oblique trees instead of axis-aligned trees | Gradient boosting forests · pendigits | test error (%): 2.87±0.01 -> 2.00±0.04 | Helped |
| Use globally optimized oblique trees instead of axis-aligned trees | Gradient boosting forests · CIFAR100 | test error (%): 26.71±0.05 -> 26.64±0.02 | Helped |
| Use globally optimized oblique trees instead of axis-aligned trees | Gradient boosting forests · news20 | test error (%): 19.78±0.00 -> 16.65±0.04 | Helped |
| Use globally optimized oblique trees instead of axis-aligned trees | Gradient boosting forests · real-sim | test error (%): 3.05±0.00 -> 2.12±0.02 | Helped |
| Use globally optimized oblique trees instead of axis-aligned trees | Gradient boosting forests · cpuact | test RMSE: 2.25±0.00 -> 2.23±0.02 | Helped |
| Use globally optimized oblique trees instead of axis-aligned trees | Gradient boosting forests · CT-slice | test RMSE: 1.24±0.00 -> 0.90±0.02 | Helped |
| Use globally optimized oblique trees instead of axis-aligned trees | Gradient boosting forests · casp | test RMSE: 3.46±0.00 -> 3.39±0.01 | Helped |
| Use globally optimized oblique trees instead of axis-aligned trees | Gradient boosting forests · superconduct | test RMSE: 8.73±0.00 -> 8.68±0.02 | Helped |
| Use globally optimized oblique trees instead of axis-aligned trees | Gradient boosting forests · year | test RMSE: 8.91±0.00 -> 8.73±0.01 | Helped |
| Limit TAO training to 3 iterations (ablation) | GB-TAO · classification datasets in Figure 3 | test error | No clear effect |
Closest papers by meaning
- Minimax Rates and Spectral Distillation for Tree Ensembles (NeurIPS 2026 · similarity 0.87)
- Free Lunch in the Forest: Functionally-Identical Pruning of Boosted Tree Ensembles (AAAI 2025 · similarity 0.87)
- Gradient Regularized Newton Boosting Trees with Global Convergence (NeurIPS 2026 · similarity 0.87)
- NRGBoost: Energy-Based Generative Boosted Trees (ICLR 2025 · similarity 0.86)
- Generalized additive models via direct optimization of regularized decision stump forests (ICML 2025 · similarity 0.86)
- GRANDE: Gradient-Based Decision Tree Ensembles for Tabular Data (ICLR 2024 · similarity 0.86)
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.