AI Paper Atlas

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

AnoShift: A Distribution Shift Benchmark for Unsupervised Anomaly Detection

Marius Dragoi, Elena Burceanu, Emanuela Haller, Andrei Manolache, Florin Brad

NeurIPS 2022 · Label-efficient and robust learning › OOD and anomaly detection · approach: Anomaly detection with contaminated data · 15 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
Evaluate on chronologically farther data rather than IID test dataUnsupervised anomaly detectors on Kyoto-2006+ · AnoShift IID to NEAR to FARROC-AUC: 92.67 -> 87.00 -> 34.53 (deepSVDD)Hurt
Use teacher-guided knowledge distillation on successive yearly data instead of IID training (ablation)BERT for anomalies · Kyoto-2006+ 2014ROC-AUC: 28.56 -> 43.65Improved
Use teacher-guided knowledge distillation on successive yearly data instead of finetuning (ablation)BERT for anomalies · Kyoto-2006+ 2015ROC-AUC: 53.39 -> 59.93Improved
Use teacher-guided knowledge distillation on successive yearly data instead of IID training (ablation)BERT for anomalies · Kyoto-2006+ 2015ROC-AUC: 49.01 -> 59.93Improved
Use teacher-guided distillation rather than IID training (ablation)BERT for anomalies · Kyoto-2006+ 2011 and 2012ROC-AUC: 88.95 -> 83.74; 95.85 -> 95.10Mixed
Do not binarize numerical features (ablation)IsoForest · Kyoto-2006+ FARROC-AUC: 27.16 -> 32.81Improved
Do not binarize numerical features (ablation)ECOD · Kyoto-2006+ FARROC-AUC: 49.19 -> 60.84Improved
Do not binarize numerical features (ablation)COPOD · Kyoto-2006+ FARROC-AUC: 50.42 -> 60.12Improved
Do not binarize numerical features (ablation)AE for anomalies · Kyoto-2006+ FARROC-AUC: 19.96 -> 30.79Improved
Do not binarize numerical features (ablation)OC-SVM · Kyoto-2006+ IID / NEAR / FARROC-AUC: 76.86 -> 81.87; 71.43 -> 71.24; 49.57 -> 50.95Mixed
Do not binarize numerical features (ablation)IsoForest · Kyoto-2006+ IID / NEAR / FARROC-AUC: 86.09 -> 94.41; 75.26 -> 95.13; 27.16 -> 32.81Improved
Do not binarize numerical features (ablation)ECOD · Kyoto-2006+ IID / NEAR / FARROC-AUC: 84.76 -> 79.38; 44.87 -> 69.80; 49.19 -> 60.84Mixed

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