Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes
DIBS: Diversity Inducing Information Bottleneck in Model Ensembles
Read the paperCheck 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 DIBS instead of Random or ABE | 4-layer CNN ensemble · MNIST translation, rotation, interpolation-extrapolation, and color shifts | accuracy | Helped |
| Use DIBS instead of Random or ABE | 4-layer CNN ensemble · MNIST to SVHN transfer | accuracy | Helped |
| Use DIBS instead of Random, NCP, or ABE | ensemble OOD detector · CIFAR-10 and CIFAR-100 with ImageNet, Gaussian, and Uniform OOD inputs | AUROC and AUPR | Helped |
| Vary the number of decoder heads K (ablation) | DIBS on TinyImageNet with VGG · TinyImageNet classification | accuracy | No clear effect |
| Use 5 DIBS ensemble models instead of baseline ensemble sizes (ablation) | TinyImageNet classification ensemble · TinyImageNet classification | accuracy | Helped |
Closest papers by meaning
- Maximizing Overall Diversity for Improved Uncertainty Estimates in Deep Ensembles (AAAI 2020 · similarity 0.91)
- Diversity Matters When Learning From Ensembles (NeurIPS 2021 · similarity 0.90)
- Neural Ensemble Search for Uncertainty Estimation and Dataset Shift (NeurIPS 2021 · similarity 0.89)
- Improving Adversarial Robustness via Promoting Ensemble Diversity (ICML 2019 · similarity 0.89)
- On the Diversity of Adversarial Ensemble Learning (ICML 2025 · similarity 0.88)
- Ensemble Pruning for Out-of-distribution Generalization (ICML 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.