Skip to Main content Skip to Navigation
Conference papers

Anomaly Detection With Conditional Variational Autoencoders

Abstract : Exploiting the rapid advances in probabilistic inference, in particular variational Bayes and variational au-toencoders (VAEs), for anomaly detection (AD) tasks remains an open research question. Previous works argued that training VAE models only with inliers is insufficient and the framework should be significantly modified in order to discriminate the anomalous instances. In this work, we exploit the deep conditional variational autoencoder (CVAE) and we define an original loss function together with a metric that targets hierarchically structured data AD. Our motivating application is a real world problem: monitoring the trigger system which is a basic component of many particle physics experiments at the CERN Large Hadron Collider (LHC). In the experiments we show the superior performance of this method for classical machine learning (ML) benchmarks and for our application.
Complete list of metadata

Cited literature [31 references]  Display  Hide  Download

https://hal.inria.fr/hal-02396279
Contributor : Cecile Germain <>
Submitted on : Thursday, December 5, 2019 - 8:53:23 PM
Last modification on : Thursday, July 8, 2021 - 3:50:39 AM
Long-term archiving on: : Friday, March 6, 2020 - 6:10:15 PM

File

paper.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02396279, version 1

Citation

Adrian Pol, Victor Berger, Gianluca Cerminara, Cécile Germain, Maurizio Pierini. Anomaly Detection With Conditional Variational Autoencoders. ICMLA 2019 - 18th IEEE International Conference on Machine Learning and Applications, Dec 2019, Boca Raton, United States. ⟨hal-02396279⟩

Share

Metrics

Record views

257

Files downloads

5224