Scenario-based Generalization bound for Anomaly Detection Support Vector Machine Ensembles

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

This work proposes an ensemble of robust Support-Vector-Machine (SVM) classifiers to monitor systems health states given uncertain measurements from multiple sensors. Scenario optimization is a well-established theory to
solve optimization problems in the presence of uncertainty and it is used to render a formal bound on the SVM
misclassification probability. This probabilistic certificate of generalization (robustness) holds non-asymptotically
and for any stationary random mechanism generating the data. A novel selection strategy is introduced seeking
an ensemble design that maximizes accuracy in the prediction and robustness given by Scenario theory. The
framework is tested on a Prognostics and Health Management (PHM) challenge problem launched by the ARAMIS
group in 2020 where a set of sensor measurements and labels are provided to abnormal operational states. The points
of strength and the weaknesses of the proposed framework are presented and discussed.
Original languageEnglish
Title of host publicationProceedings of the 30th European Safety and Reliability Conference and the 15th Probabilistic Safety Assessment and Management Conference
EditorsPiero Baraldi, Francesco Di Maio, Enrico Zio
ISBN (Electronic)978-981-14-8593-0
Publication statusPublished - 1 Nov 2020

Keywords

  • Scenario Optimization
  • Anomaly Detection
  • Generalization learning
  • lifetime analysis
  • Prognostic and health management
  • Support vector machine (SVM)

Fingerprint Dive into the research topics of 'Scenario-based Generalization bound for Anomaly Detection Support Vector Machine Ensembles'. Together they form a unique fingerprint.

Cite this