Change Detection in Multivariate Datastreams Controlling False Alarms
Abstract
We introduce QuantTree Exponentially Weighted Moving Average (QT-EWMA), a novel change-detection algorithm for multivariate datastreams that can operate in a nonparametric and online manner. QT-EWMA can be configured to yield a target Average Run Length (ARL $_0$ 0 ), thus controlling the expected time before a false alarm. Control over false alarms has many practical implications and is rarely guaranteed by online change-detection algorithms that can monitor multivariate datastreams whose distribution is unknown. Our experiments, performed on synthetic and real-world datasets, demonstrate that QT-EWMA controls the ARL $_0$ 0 and the false alarm rate better than state-of-the-art methods operating in similar conditions, achieving comparable detection delays.
Cite
Text
Frittoli et al. "Change Detection in Multivariate Datastreams Controlling False Alarms." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2021. doi:10.1007/978-3-030-86486-6_26Markdown
[Frittoli et al. "Change Detection in Multivariate Datastreams Controlling False Alarms." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2021.](https://mlanthology.org/ecmlpkdd/2021/frittoli2021ecmlpkdd-change/) doi:10.1007/978-3-030-86486-6_26BibTeX
@inproceedings{frittoli2021ecmlpkdd-change,
title = {{Change Detection in Multivariate Datastreams Controlling False Alarms}},
author = {Frittoli, Luca and Carrera, Diego and Boracchi, Giacomo},
booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
year = {2021},
pages = {421-436},
doi = {10.1007/978-3-030-86486-6_26},
url = {https://mlanthology.org/ecmlpkdd/2021/frittoli2021ecmlpkdd-change/}
}