Research map / Architectures and efficiency / RNNs and normalization
Multi-Layered Gradient Boosting Decision Trees
Ji Feng, Yang Yu, Zhi-Hua Zhou
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 multi-layered GBDT instead of XGBoost | Income classification · Income Dataset | accuracy: .8719 -> .8742 | Improved |
| Use multi-layered GBDT instead of XGBoost Stacking | Income classification · Income Dataset | accuracy: .8697 -> .8742 | Improved |
| Use multi-layered GBDT instead of NN BackProp | Income classification · Income Dataset | accuracy: .8534 -> .8742 | Improved |
| Use multi-layered GBDT instead of XGBoost | Protein classification · Protein Dataset | accuracy: .5937 $\pm$ .0324 -> .5948 $\pm$ .0268 | Mixed |
| Use multi-layered GBDT instead of XGBoost Stacking | Protein classification · Protein Dataset | accuracy: .5592 $\pm$ .0400 -> .5948 $\pm$ .0268 | Improved |
| Use multi-layered GBDT instead of NN BackProp | Protein classification · Protein Dataset | accuracy: .5907 $\pm$ .0268 -> .5948 $\pm$ .0268 | Mixed |
| Use one intermediate layer instead of two (ablation) | mGBDT · Protein Dataset | accuracy: .5948 $\pm$ .0268 -> .6160 $\pm$ .0323 | Improved |
| Add one intermediate layer (ablation) | mGBDT · Protein Dataset | accuracy: .5948 $\pm$ .0268 -> .5897 $\pm$ .0312 | Hurt |
| Add two intermediate layers (ablation) | mGBDT · Protein Dataset | accuracy: .5948 $\pm$ .0268 -> .5782 $\pm$ .0229 | Hurt |
| Increase depth from one to four intermediate layers | NN TargetProp · Protein Dataset | accuracy: .5964 -> .3654 | Hurt |
Closest papers by meaning
- Boost then Convolve: Gradient Boosting Meets Graph Neural Networks (ICLR 2021 · similarity 0.89)
- GradTree: Learning Axis-Aligned Decision Trees with Gradient Descent (AAAI 2024 · similarity 0.85)
- Gradient Regularized Newton Boosting Trees with Global Convergence (NeurIPS 2026 · similarity 0.85)
- NRGBoost: Energy-Based Generative Boosted Trees (ICLR 2025 · similarity 0.85)
- Symbolic Regression Enhanced Decision Trees for Classification Tasks (AAAI 2024 · similarity 0.83)
- Better by default: Strong pre-tuned MLPs and boosted trees on tabular data (NeurIPS 2024 · similarity 0.83)
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.
