AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Learning Credal Ensembles via Distributionally Robust Optimization

Kaizheng Wang, Ghifari Adam Faza, Fabio Cuzzolin, Siu Lun Chau, David Moens, Hans Hallez

ICML 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Confidence calibration objectives · 0 citations (OpenAlex)

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 changedWhereMeasuredEffect
use box credal set instead of convex hull (ablation)CreDRO · CIFAR10 vs SVHN, M=5AUROC: 96.0 -> 96.6Improved
use box credal set instead of convex hull (ablation)CreDRO · CIFAR10 vs SVHN, M=20AUROC: 97.2 -> 97.4Improved
increase ensemble size from 5 to 20 (ablation)CreDRO · CIFAR10 vs SVHNAUROC: 96.6 -> 97.4Improved
vary delta_G from 0.5 to 0.9 (ablation)CreDRO · CIFAR10 vs SVHNAUROC: 96.6 -> 96.1No clear effect
add top-loss selection instead of random upweighting at 10% label noise (ablation)CreDRO · CIFAR10 vs SVHN, 10% noiseAUROC: 88.7 -> 92.2Improved
add top-loss selection instead of random upweighting at 20% label noise (ablation)CreDRO · CIFAR10 vs SVHN, 20% noiseAUROC: 83.8 -> 88.9Improved
add DRO training to a classical ensemble (EN-DRO vs DE)Deep ensemble · CIFAR10 vs SVHN, M=20AUROC: 94.8 -> 95.7Improved
add varied DRO levels and credal set (CreDRO vs EN-DRO)EN-DRO · CIFAR10 vs SVHN, M=20AUROC: 95.7 -> 97.4Improved
use CreDRO averaged prediction instead of DEDeep ensemble · CIFAR10 testaccuracy: 0.9569 -> 0.9637Improved
use CreDRO averaged prediction instead of DEDeep ensemble · CIFAR10 testECE: 0.0051 -> 0.0038Improved
use DRO training instead of ERM (CreDRO vs CreWra)CreWra · CIFAR10 training timeseconds: 6259.74 -> 6567.93Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy