Research map / Learning theory and algorithms / Prediction and decision losses
DivBO: Diversity-aware CASH for Ensemble Learning
Yu Shen, Yupeng Lu, Yang Li, Yaofeng Tu, Wentao Zhang, Bin CUI
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 diversity-aware surrogate to BO-ES (DivBO) | BO-ES · 15 datasets | average rank (test error): 3.13 -> 1.73 | Improved |
| DivBO vs strongest baseline RB-ES | post-hoc ensemble · 15 datasets | average rank (test error): 2.90 -> 1.73 | Improved |
| DivBO vs RB-ES statistical test | post-hoc ensemble · 15 datasets | datasets better/same/worse: 8 better, 4 same, 3 worse | Mixed |
| DivBO vs RS-ES statistical test | post-hoc ensemble · 15 datasets | datasets better/same/worse: 13 better, 2 same, 0 worse | Improved |
| DivBO vs BO-ES statistical test | post-hoc ensemble · 15 datasets | datasets better/same/worse: 12 better, 1 same, 2 worse | Improved |
| fix diversity weight at 0.1 instead of scheduling (ablation) | DivBO · pol | test error (%): 1.34 -> 1.66 | Hurt |
| fix diversity weight at 0.1 instead of scheduling (ablation) | DivBO · quake | test error (%): 45.55 -> 45.63 | No clear effect |
| fix diversity weight at 0.1 instead of scheduling (ablation) | DivBO · house_8L | test error (%): 10.80 -> 11.51 | Hurt |
| use bagging instead of ensemble selection (ablation) | DivBO · elevators | test error (%): 9.40 -> 10.98 | Hurt |
| use stacking instead of ensemble selection (ablation) | DivBO · pol | test error (%): 1.34 -> 1.54 | Hurt |
| increase ensemble size from 25 to 100 (ablation) | BO-ES · quake | test error (%): 46.10 -> 47.43 | Hurt |
| increase ensemble size from 25 to 100 (ablation) | BO-ES · wind | test error (%): 14.04 -> 14.04 | No clear effect |
Closest papers by meaning
- PSEO: Optimizing Post-hoc Stacking Ensemble Through Hyperparameter Tuning (AAAI 2026 · similarity 0.88)
- Diverse Ensemble Evolution: Curriculum Data-Model Marriage (NeurIPS 2018 · similarity 0.86)
- Diversified Batch Selection for Training Acceleration (ICML 2024 · similarity 0.86)
- Sharpness-diversity tradeoff: improving flat ensembles with SharpBalance (NeurIPS 2024 · similarity 0.86)
- Put CASH on Bandits: A Max K-Armed Problem for Automated Machine Learning (NeurIPS 2025 · similarity 0.84)
- GRANDE: Gradient-Based Decision Tree Ensembles for Tabular Data (ICLR 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.
