AI Paper Atlas

Research map / Label-efficient and robust learning / OOD and anomaly detection

Igeood: An Information Geometry Approach to Out-of-Distribution Detection

Eduardo Dadalto Camara Gomes, Florence Alberge, Pierre Duhamel, Pablo Piantanida

ICLR 2022 · Label-efficient and robust learning › OOD and anomaly detection · approach: Spurious-aware OOD detection · 11 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 Fisher-Rao distances between softmax outputs and class-conditional centroids instead of EnergyBlack-Box OOD detection · Six model/dataset contexts, eight OOD datasets (average)TNR at TPR-95%: 51.1 -> 51.6Improved
Use Fisher-Rao distances between softmax outputs and class-conditional centroids instead of EnergyBlack-Box OOD detection · Six model/dataset contexts, eight OOD datasets (average)AUROC: 85.2 -> 85.7Improved
Combine logits and low-level feature scores, including OOD-feature statisticsWhite-Box Igeood+ · Validation on OOD data, average across models and in-distribution datasetsTNR at TPR-95%: 81.1 -> 92.9Improved
Combine logits and low-level feature scores, including OOD-feature statisticsWhite-Box Igeood+ · Validation on OOD data, average across models and in-distribution datasetsAUROC: 94.8 -> 98.5Improved
Combine logits and low-level feature scores, tuned using adversarial dataWhite-Box Igeood · Validation on adversarial data, average across models and in-distribution datasetsTNR at TPR-95%: 77.0 -> 79.5Improved
Combine logits and low-level feature scores, tuned using adversarial dataWhite-Box Igeood · Validation on adversarial data, average across models and in-distribution datasetsAUROC: 92.8 -> 93.4Improved
Add intermediate-feature scores to the logits score (ablation)White-Box Igeood · OOD detectionOOD detection performanceImproved
Add OOD-feature statistics to the logits and intermediate-feature scores (ablation)White-Box Igeood+ · OOD detection with validation OOD dataOOD detection performanceImproved
Replace the minimum class-centroid logit distance with the sum across class centroids (ablation)Logits-level Igeood · OOD detectionOOD detection performanceImproved
Replace the minimum class-conditional feature distance with the sum across classes (ablation)Feature-level Igeood · OOD detectionOOD detection performanceNo clear effect
Use input preprocessing to perturb the test input in the direction that increases the Igeood score (ablation)Grey-Box Igeood · Grey-Box OOD detectionTNR at TPR-95%Mixed

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