Research map / Architectures and efficiency / RNNs and normalization
Boosting Ensemble Accuracy by Revisiting Ensemble Diversity Metrics
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 focal-model negative sampling, equal-size team comparisons, and clustering instead of the corresponding Q-diversity selection | Ensemble selection from the ImageNet base-model pool · ImageNet | ensemble accuracy lower bound (%): 70.79% -> 73.26% | Helped |
| Use FQ diversity selection instead of the corresponding Q-diversity metrics | Ensemble selection from the CIFAR-10 base-model pool · CIFAR-10 | ensemble accuracy lower bound (%): 93.56% -> 94.40% | Helped |
| Fuse FQ-BD, FQ-KW, and FQ-GD selections with EQ-diversity | FQ ensemble selection · CIFAR-10 | ensemble accuracy average (%): EQ-diversity with BD+KW+GD improved the average ensemble accuracy to 96.40%. | Helped |
| Fuse FQ-BD, FQ-KW, and FQ-GD selections with EQ-diversity | FQ ensemble selection · ImageNet | ensemble accuracy average (%): EQ-diversity with BD+KW+GD improved the ensemble accuracy average to 79.52%. | Helped |
| Compare only ensemble teams of equal size rather than teams of different sizes together (ablation) | Q-diversity-based ensemble selection · CIFAR-10 | correlation between diversity and ensemble accuracy | Helped |
| Use focal-model negative samples rather than one shared pool of negative samples (ablation) | Q-diversity-based ensemble selection · CIFAR-10 | correlation between diversity and ensemble accuracy | Helped |
| Apply binary clustering with strategic centroid initialization to set the diversity threshold (ablation) | FQ diversity selection · CIFAR-10 | quality of selected ensemble teams | Helped |
Closest papers by meaning
- Diversity Matters When Learning From Ensembles (NeurIPS 2021 · similarity 0.90)
- Ensembles of Locally Independent Prediction Models (AAAI 2020 · similarity 0.88)
- Neural Ensemble Search for Uncertainty Estimation and Dataset Shift (NeurIPS 2021 · similarity 0.87)
- On the Diversity of Adversarial Ensemble Learning (ICML 2025 · similarity 0.87)
- Sharpness-diversity tradeoff: improving flat ensembles with SharpBalance (NeurIPS 2024 · similarity 0.87)
- Theoretical Limitations of Ensembles in the Age of Overparameterization (ICML 2025 · 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.