AI Paper Atlas

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

Invariant Anomaly Detection under Distribution Shifts: A Causal Perspective

João B. S. Carvalho, Mengtao Zhang, Robin Geyer, Carlos Cotrini, Joachim M. Buhmann

NeurIPS 2023 · Label-efficient and robust learning › OOD and anomaly detection · approach: Anomaly detection with contaminated data · 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
add PCIR regularizerMeanShift · Waterbirds out-of-distributionAUROC: 0.629 -> 0.726Improved
add PCIR regularizerReverseDistil · Waterbirds out-of-distributionAUROC: 0.617 -> 0.723Improved
add PCIR regularizerWaterbirds shortcut setting, all six detectors · Waterbirds in-distributionAUROC: STFPM 0.699 -> 0.724; CFA 0.752 -> 0.785Improved
add PCIR regularizerMeanShift · Camelyon17 (3 train envs), OOD e5AUROC: 0.690 -> 0.731Improved
add PCIR regularizerRed PANDA · Camelyon17 (2 train envs), OOD e5AUROC: 0.651 -> 0.740Improved
add PCIR regularizerCFA · Camelyon17 (3 train envs)AUROC: 0.643 -> 0.644No clear effect
add PCIR regularizerReverseDistil · Camelyon17 (2 train envs), OOD e5AUROC: 0.723 -> 0.718Hurt
add PCIR regularizerMeanShift · MNIST, 4 covariate shiftsAUROC: 0.701 -> 0.770Improved
add PCIR regularizerSTFPM · Fashion-MNIST, 1 covariate shiftAUROC: 0.530 -> 0.601Improved
add PCIR regularizerSTFPM · Fashion-MNIST, in-distributionAUROC: 0.752 -> 0.733Hurt
add PCIR regularizerRed PANDA · Fashion-MNIST, in-distributionAUROC: 0.729 -> 0.715Hurt
add PCIR regularizer across all methods and OOD environmentssix anomaly detectors · all benchmarksmean-AUROC gain: 2.5% to 20%Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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