Research map / Large language models / AI and society
A Unified Framework for Knowledge Intensive Gradient Boosting: Leveraging Human Experts for Noisy Sparse Domains
Harsha Kokel, Phillip Odom, Shuo Yang, Sriraam Natarajan
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 |
|---|---|---|---|
| add monotonic advice penalty (KiGB) to gradient boosting | scikit-learn GB · Cleveland | accuracy: 0.677 -> 0.737 | Improved |
| add monotonic advice penalty (KiGB) to gradient boosting | scikit-learn GB · Autompg | MSE: 13.623 -> 9.793 | Improved |
| add monotonic advice penalty (KiGB) to gradient boosting | scikit-learn GB · Boston | MSE: 21.493 -> 24.065 | Hurt |
| add monotonic advice penalty (KiGB) to gradient boosting | scikit-learn GB · Adult | accuracy: 0.853 -> 0.855 | No clear effect |
| add monotonic advice penalty (KiGB) to gradient boosting | LightGBM · Logistics | MSE: 1.898 -> 1.851 | Improved |
| add monotonic advice penalty (KiGB) to gradient boosting | scikit-learn GB · HELOC | accuracy: 0.7 -> 0.717 | Improved |
| use KiGB soft advice penalty instead of Monoensemble | scikit-learn GB · Car | accuracy: 0.765 -> 0.984 | Improved |
| use KiGB soft advice penalty instead of Monoensemble | scikit-learn GB · Australian | accuracy: 0.884 vs 0.855 | Hurt |
| use KiGB advice penalty instead of LightGBM monotonic constraints | LightGBM · Boston | MSE: 16.292 -> 15.496 | Improved |
| use KiGB advice penalty instead of LightGBM monotonic constraints | LightGBM · Whitewine | MSE: 0.467 -> 0.45 | Improved |
| use KiGB advice penalty instead of LightGBM monotonic constraints | LightGBM · Adult | accuracy: 0.863 -> 0.865 | No clear effect |
| increase advice weight lambda with negative epsilon (strict margin) (ablation) | KiGB · hyperparameter study | accuracy / MSE | Hurt |
Closest papers by meaning
- Active Learning for Domain Adaptation: An Energy-Based Approach (AAAI 2022 · similarity 0.86)
- Gradient Boosting Performs Gaussian Process Inference (ICLR 2023 · similarity 0.85)
- Learning Gradient Descent: Better Generalization and Longer Horizons (ICML 2017 · similarity 0.85)
- Statistical Inference for Gradient Boosting Regression (NeurIPS 2025 · similarity 0.84)
- Category-Aware Active Domain Adaptation (ICML 2024 · similarity 0.83)
- Self-Training with Dynamic Weighting for Robust Gradual Domain Adaptation (NeurIPS 2025 · 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.
