AI Paper Atlas

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

Real-Time Nonparametric Anomaly Detection in High-Dimensional Settings

Mehmet Necip Kurt, Yasin Yılmaz, Xiaodong Wang

TPAMI 2020 · Label-efficient and robust learning › OOD and anomaly detection · approach: Anomaly detection with contaminated data

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 the GEM-based CUSUM-like detector instead of benchmark testsGEM-based detector · IEEE-57 smart grid FDI attackdetection delay vs false alarm periodImproved
use proposed detectors instead of benchmark testsGEM- and PCA-based detectors · HAPT activity changedetection delay vs false alarm periodImproved
use p-value based statistic instead of raw NN distance (ODIT)GEM-based detector · N-BaIoT thermostat BASHLITE spamfalse alarm rateImproved
use p-value statistic instead of mean-deviation nonparametric CUSUMproposed detectors · N-BaIoT thermostat BASHLITE spamdetection delayImproved
apply GEM detector after PCA projection to low dimensionGEM-based detector · HAPT and N-BaIoTcomputational costImproved
use proposed approximation instead of Wald's approximation for false alarm periodCUSUM-like detector · Monte Carlo alpha=0.1average false alarm periodImproved
decrease alpha (ablation)CUSUM-like detector · Monte Carlofalse alarm period and detection delayMixed
increase attack magnitudeGEM-based detector · IEEE-57 smart grid FDI attackdetection delayImproved

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