skip to main content

This content will become publicly available on March 7, 2016

Title: Problematic projection to the in-sample subspace for a kernelized anomaly detector

We examine the properties and performance of kernelized anomaly detectors, with an emphasis on the Mahalanobis-distance-based kernel RX (KRX) algorithm. Although the detector generally performs well for high-bandwidth Gaussian kernels, it exhibits problematic (in some cases, catastrophic) performance for distances that are large compared to the bandwidth. By comparing KRX to two other anomaly detectors, we can trace the problem to a projection in feature space, which arises when a pseudoinverse is used on the covariance matrix in that feature space. Here, we show that a regularized variant of KRX overcomes this difficulty and achieves superior performance over a wide range of bandwidths.
 [1] ;  [1]
  1. Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
Publication Date:
OSTI Identifier:
Report Number(s):
Journal ID: ISSN 1545-598X
Grant/Contract Number:
Accepted Manuscript
Journal Name:
IEEE Geoscience and Remote Sensing Letters
Additional Journal Information:
Journal Volume: 13; Journal Issue: 4; Journal ID: ISSN 1545-598X
Research Org:
Los Alamos National Lab. (LANL), Los Alamos, NM (United States)
Sponsoring Org:
Country of Publication:
United States
58 GEOSCIENCES; 97 MATHEMATICS AND COMPUTING; anomaly detection; kernel density estimation; mahalanobis distance; kernel-RX; adaptive signal detection; algorithms; covariance matrices; data models; detectors; multidimensional signal processing; pattern recognition; remote sensing; singular value decomposition; spectral analysis