Likelihood-Based Unsupervised Anomaly Detection in CMS Dijet Events
Abstract
We present an unsupervised search for anomalous dijet events in proton--proton collision data using neural spline flow density estimation. A normalizing flow model is trained on a high-dimensional feature space comprising jet, dijet, and event-level observables to learn the dominant Standard Model background directly from data, without assuming a specific signal hypothesis. Events assigned low likelihood under the learned density are identified as potential anomalous events. Using this approach on a CMS Open Data dijet sample, we investigate extreme events in the tail of the anomaly-score distribution and perform an extensive validation and robustness study. This study includes feature-level statistical comparisons, mass decorrelation tests, permutation-based null tests, and evaluations of training stability. The selected anomalous events exhibit notable departures from the background-only expectation, primarily in jet-substructure observables, while remaining stable under several known sources of bias in unsupervised learning. The identified anomalies are distributed across the kinematic phase space and do not exhibit a narrow structure in the dijet invariant-mass spectrum. Instead, they show correlated deviations across multiple observables, consistent with a multivariate difference in jet substructure and event topology rather than a localized resonance. While no claim of new physics is made, this study demonstrates that neural spline flow-based density estimation can be sensitive to rare, structured deviations in collider data and may provide a model-independent exploratory tool for searches for physics beyond the Standard Model at the LHC.
Related
- Pareto-optimal quantum kernel selection for unsupervised anomaly detection on real malware beaconing data
- DANTE: A Reference-Guided Unsupervised Pipeline for Extended-Transient Anomaly Characterization in LIGO O4a
- MIRTO: a registration-gated, multiverse-tested evaluation protocol for unsupervised anomaly segmentation in brain MRI
- Reconstruction of Shower-like Events in NEON Using Likelihood and Graph Neural Network Methods
- Performance of a Gamma Detection Unit with Likelihood-Based Localization Method
BibTeX
@misc{chebrolu2026likelihoodbased,
title = {{Likelihood-Based Unsupervised Anomaly Detection in CMS Dijet Events}},
author = {Bhavishya Chebrolu and Hitesh Rasineni and Prajwal Aaryan Immadi},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.06686},
eprint = {2609.06686},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.06686},
}
RIS
TY - UNPB
TI - Likelihood-Based Unsupervised Anomaly Detection in CMS Dijet Events
AU - Bhavishya Chebrolu
AU - Hitesh Rasineni
AU - Prajwal Aaryan Immadi
PY - 2026
DA - 2026/09/06
PB - arXiv
DO - 10.48550/arxiv.2609.06686
UR - https://arxiv.org/abs/2609.06686
AB - We present an unsupervised search for anomalous dijet events in proton--proton collision data using neural spline flow density estimation. A normalizing flow model is trained on a high-dimensional feature space comprising jet, dijet, and event-level observables to learn the dominant Standard Model background directly from data, without assuming a specific signal hypothesis. Events assigned low likelihood under the learned density are identified as potential anomalous events. Using this approach on a CMS Open Data dijet sample, we investigate extreme events in the tail of the anomaly-score distribution and perform an extensive validation and robustness study. This study includes feature-level statistical comparisons, mass decorrelation tests, permutation-based null tests, and evaluations of training stability. The selected anomalous events exhibit notable departures from the background-only expectation, primarily in jet-substructure observables, while remaining stable under several known sources of bias in unsupervised learning. The identified anomalies are distributed across the kinematic phase space and do not exhibit a narrow structure in the dijet invariant-mass spectrum. Instead, they show correlated deviations across multiple observables, consistent with a multivariate difference in jet substructure and event topology rather than a localized resonance. While no claim of new physics is made, this study demonstrates that neural spline flow-based density estimation can be sensitive to rare, structured deviations in collider data and may provide a model-independent exploratory tool for searches for physics beyond the Standard Model at the LHC.
ER -