AI Paper Atlas

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

Transfer Learning for Anomaly Detection through Localized and Unsupervised Instance Selection

Vercruyssen Vincent, Meert Wannes, Davis Jesse

AAAI 2020 · Label-efficient and robust learning › OOD and anomaly detection · approach: Anomaly detection with contaminated data · 31 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 SSKNNO instead of KNN as target classifier (ablation)CORAL transfer · Anomaly transfer benchmarkaverage AUROC: 0.666 -> 0.735Improved
use SSKNNO instead of KNN as target classifier (ablation)TransferAll · Anomaly transfer benchmarkaverage AUROC: 0.649 -> 0.727Improved
use SSKNNO instead of KNN as target classifier (ablation)GFK transfer · Anomaly transfer benchmarkaverage AUROC: 0.642 -> 0.705Improved
use SSKNNO instead of KNN as target classifier (ablation)CBIT transfer · Anomaly transfer benchmarkaverage AUROC: 0.623 -> 0.713Improved
use SSKNNO instead of KNN as target classifier (ablation)TCA transfer · Anomaly transfer benchmarkaverage AUROC: 0.656 -> 0.533Hurt
replace global domain alignment/transfer-all with localized instance selection (LocIT transfer) under same SSKNNO classifier (ablation)SSKNNO · Anomaly transfer benchmarkaverage AUROC: 0.735 -> 0.762Improved
add transferred labeled source instances (LocIT) over unsupervised detectionanomaly detector · Anomaly transfer benchmarkaverage AUROC: 0.705 -> 0.762Improved
increase neighborhood size psi to >= 20 (ablation)LocIT · Anomaly transfer benchmarkAUROCImproved
increase fraction of labeled source instances (ablation)LocIT · Anomaly transfer benchmarkAUROCImproved
LocIT transfer vs KNNO on real water usage dataLocIT · Water usage, 6 store pairsaverage AUROC: 0.754 -> 0.779Mixed

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