Bump Hunting in Latent Space

Unsupervised anomaly detection could be crucial in future analyses searching for rare phenomena in large datasets, as for example collected at the LHC. To this end, we introduce a physics inspired variational autoencoder (VAE) architecture which performs competitively and robustly on the LHC Olympic...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Bortolato, Blaž, Dillon, Barry M, Kamenik, Jernej F, Smolkovič, Aleks
Format Paper Journal Article
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 07.06.2022
Subjects
Anomalies
Computer Science - Learning
Datasets
Machine learning
Physics - High Energy Physics - Experiment
Physics - High Energy Physics - Phenomenology
Online AccessGet full text

Cover

More Information
Summary:Unsupervised anomaly detection could be crucial in future analyses searching for rare phenomena in large datasets, as for example collected at the LHC. To this end, we introduce a physics inspired variational autoencoder (VAE) architecture which performs competitively and robustly on the LHC Olympics Machine Learning Challenge datasets. We demonstrate how embedding some physical observables directly into the VAE latent space, while at the same time keeping the classifier manifestly agnostic to them, can help to identify and characterise features in measured spectra as caused by the presence of anomalies in a dataset.
ISSN:2331-8422
DOI:10.48550/arxiv.2103.06595