Toward supervised anomaly detection
From MaRDI portal
Abstract: Anomaly detection is being regarded as an unsupervised learning task as anomalies stem from adversarial or unlikely events with unknown distributions. However, the predictive performance of purely unsupervised anomaly detection often fails to match the required detection rates in many tasks and there exists a need for labeled data to guide the model generation. Our first contribution shows that classical semi-supervised approaches, originating from a supervised classifier, are inappropriate and hardly detect new and unknown anomalies. We argue that semi-supervised anomaly detection needs to ground on the unsupervised learning paradigm and devise a novel algorithm that meets this requirement. Although being intrinsically non-convex, we further show that the optimization problem has a convex equivalent under relatively mild assumptions. Additionally, we propose an active learning strategy to automatically filter candidates for labeling. In an empirical study on network intrusion detection data, we observe that the proposed learning methodology requires much less labeled data than the state-of-the-art, while achieving higher detection accuracies.
Recommendations
Cited in
(15)- Anomaly and novelty detection for robust semi-supervised learning
- Interpreting rate-distortion of variational autoencoder and using model uncertainty for anomaly detection
- Anomaly detection with inexact labels
- Time series anomaly detection based on shapelet learning
- Semi-supervised novelty detection
- Loda: lightweight on-line detector of anomalies
- Learning to control a structured-prediction decoder for detection of HTTP-layer DDoS attackers
- Security analysis of online centroid anomaly detection
- scientific article; zbMATH DE number 2215881 (Why is no real title available?)
- Semi-supervised anomaly detection in dynamic communication networks
- Almost aggregations in the gravitational clustering to perform anomaly detection
- Search-based cost-sensitive hypergraph learning for anomaly detection
- Statistical learning methods for information security: fundamentals and case studies
- Unsupervised anomaly detection using inverse generative adversarial networks
- General supervised learning framework for open world classification
This page was built for publication: Toward supervised anomaly detection
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4907836)
