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
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add PCIR regularizer | MeanShift · Waterbirds out-of-distribution | AUROC: 0.629 -> 0.726 | Improved |
| add PCIR regularizer | ReverseDistil · Waterbirds out-of-distribution | AUROC: 0.617 -> 0.723 | Improved |
| add PCIR regularizer | Waterbirds shortcut setting, all six detectors · Waterbirds in-distribution | AUROC: STFPM 0.699 -> 0.724; CFA 0.752 -> 0.785 | Improved |
| add PCIR regularizer | MeanShift · Camelyon17 (3 train envs), OOD e5 | AUROC: 0.690 -> 0.731 | Improved |
| add PCIR regularizer | Red PANDA · Camelyon17 (2 train envs), OOD e5 | AUROC: 0.651 -> 0.740 | Improved |
| add PCIR regularizer | CFA · Camelyon17 (3 train envs) | AUROC: 0.643 -> 0.644 | No clear effect |
| add PCIR regularizer | ReverseDistil · Camelyon17 (2 train envs), OOD e5 | AUROC: 0.723 -> 0.718 | Hurt |
| add PCIR regularizer | MeanShift · MNIST, 4 covariate shifts | AUROC: 0.701 -> 0.770 | Improved |
| add PCIR regularizer | STFPM · Fashion-MNIST, 1 covariate shift | AUROC: 0.530 -> 0.601 | Improved |
| add PCIR regularizer | STFPM · Fashion-MNIST, in-distribution | AUROC: 0.752 -> 0.733 | Hurt |
| add PCIR regularizer | Red PANDA · Fashion-MNIST, in-distribution | AUROC: 0.729 -> 0.715 | Hurt |
| add PCIR regularizer across all methods and OOD environments | six anomaly detectors · all benchmarks | mean-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.
- Red PANDA — Red PANDA: Disambiguating Image Anomaly Detection by Removing Nuisance Factors (ICLR 2023) · better on 3 of 3 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
- Filter or Compensate: Towards Invariant Representation from Distribution Shift for Anomaly Detection (AAAI 2025 · similarity 0.91)
- ResAD: A Simple Framework for Class Generalizable Anomaly Detection (NeurIPS 2024 · similarity 0.89)
- The Elephant in the Room: Towards A Reliable Time-Series Anomaly Detection Benchmark (NeurIPS 2024 · similarity 0.88)
- An Evidence-Based Post-Hoc Adjustment Framework for Anomaly Detection Under Data Contamination (NeurIPS 2025 · similarity 0.87)
- Long-Tailed Anomaly Detection with Learnable Class Names (CVPR 2024 · similarity 0.87)
- ADPretrain: Advancing Industrial Anomaly Detection via Anomaly Representation Pretraining (NeurIPS 2025 · similarity 0.87)
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.
