Covariant Contrastive Learning for Uncertainty-Aware Anomaly Detection

Abstract

Machine-learning-based anomaly detection (AD) offers a promising, model-agnostic alternative to traditional LHC analyses, allowing to search for many signals at once. Recent advances in representation learning motivate the use of neural embeddings to map high-dimensional physics observables into low-dimensional latent spaces better suited to statistical inference. However, the propagation of systematic uncertainty in embedded spaces remains poorly understood, severely limiting the application of these methods to real LHC analyses. We address this gap with a machine learning strategy that uses a likelihood-based regularization to improve the structure of embedded spaces. Building on previous work using supervised contrastive learning for physics-aware embeddings, our approach trains them jointly with a downstream parameterized classification task that incorporates continuous uncertainties as nuisance parameters. This method returns a latent space that covaries predictably with systematic shifts, and a model of such distortions that can be exploited in the downstream statistical test for anomaly detection. Using simulated CMS Level-1 trigger data, we show that our method successfully models continuous uncertainties in a 4D latent space with a linear parametric downstream classifier, improving both the interpretability and robustness of the statistical anomaly detection task. This framework offers a general way to study and control systematic uncertainties in latent spaces, opening the way to a new scalable statistical analysis workflow for anomaly detection at the LHC, and other high energy physics experiments.

Towards anomaly detection searches for new physics signatures including Higgs bosons with weakly supervised machine learning
Chi Lung Cheng, Julia Gonski, Julie Khalilieh Romman +5
arXiv20 Oct 2026PreprintarXiv licenceFull textHigh-energy & particle physicsarXiv
Unsupervised and probabilistic learning with Contrastive Local Learning Networks: The Restricted Kirchhoff Machine
Marcelo Guzman, Simone Ciarella, Andrea Liu
Proceedings of the National Academy of Sciences of the United States of America21 May 2026Peer-reviewedCC BY-NC-NDComputational physics Condensed matterHAL
Lyman Break Galaxy selection and redshift measurement with supervised contrastive learning
J Choppin de Janvry, C Yèche, Arjun Dey +55
HAL1 Sep 2026PreprintLicence not statedGeneral astronomy General physicsHAL
SelfTICA: contrastive learning of dynamical representations for rare-event sampling and characterization
Kai Zhu, Jintu Zhang, Pietro Novelli +2
arXiv9 Sep 2026PreprintarXiv licenceFull textPhysical chemistry Computational physicsarXiv
Anomaly detection for multijet scenarios
Gregor Kasieczka, Sung Hak Lim, Louis Moureaux +3
arXiv14 Oct 2026PreprintCC BY 4.0Full textComputational physics High-energy & particle physicsarXiv

Cite

BibTeX RIS

BibTeX

@misc{tong2026covariant,
  title = {{Covariant Contrastive Learning for Uncertainty-Aware Anomaly Detection}},
  author = {Shelley Tong and Philip Harris and Gaia Grosso},
  howpublished = {arXiv},
  year = {2026},
  month = sep,
  doi = {10.48550/arxiv.2609.35755},
  eprint = {2609.35755},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.35755},
}

RIS

TY  - UNPB
TI  - Covariant Contrastive Learning for Uncertainty-Aware Anomaly Detection
AU  - Shelley Tong
AU  - Philip Harris
AU  - Gaia Grosso
PY  - 2026
DA  - 2026/09/28
PB  - arXiv
DO  - 10.48550/arxiv.2609.35755
UR  - https://arxiv.org/abs/2609.35755
AB  - Machine-learning-based anomaly detection (AD) offers a promising, model-agnostic alternative to traditional LHC analyses, allowing to search for many signals at once. Recent advances in representation learning motivate the use of neural embeddings to map high-dimensional physics observables into low-dimensional latent spaces better suited to statistical inference. However, the propagation of systematic uncertainty in embedded spaces remains poorly understood, severely limiting the application of these methods to real LHC analyses. We address this gap with a machine learning strategy that uses a likelihood-based regularization to improve the structure of embedded spaces. Building on previous work using supervised contrastive learning for physics-aware embeddings, our approach trains them jointly with a downstream parameterized classification task that incorporates continuous uncertainties as nuisance parameters. This method returns a latent space that covaries predictably with systematic shifts, and a model of such distortions that can be exploited in the downstream statistical test for anomaly detection. Using simulated CMS Level-1 trigger data, we show that our method successfully models continuous uncertainties in a 4D latent space with a linear parametric downstream classifier, improving both the interpretability and robustness of the statistical anomaly detection task. This framework offers a general way to study and control systematic uncertainties in latent spaces, opening the way to a new scalable statistical analysis workflow for anomaly detection at the LHC, and other high energy physics experiments.
ER  -

Abstract

Related

Cite