Document Type
Article
Department/Program
Data Science
Journal Title
Machine Learning: Science and Technology
Pub Date
1-2024
Publisher
IOP Publishing
Volume
5
Issue
1
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Abstract
We introduce a physics-informed Bayesian neural network with flow-approximated posteriors using multiplicative normalizing flows for detailed uncertainty quantification (UQ) at the physics event-level. Our method is capable of identifying both heteroskedastic aleatoric and epistemic uncertainties, providing granular physical insights. Applied to deep inelastic scattering (DIS) events, our model effectively extracts the kinematic variables x, Q2, and y, matching the performance of recent deep learning regression techniques but with the critical enhancement of event-level UQ. This detailed description of the underlying uncertainty proves invaluable for decision-making, especially in tasks like event filtering. It also allows for the reduction of true inaccuracies without directly accessing the ground truth. A thorough DIS simulation using the H1 detector at HERA indicates possible applications for the future electron–ion collider. Additionally, this paves the way for related tasks such as data quality monitoring and anomaly detection. Remarkably, our approach effectively processes large samples at high rates.
Recommended Citation
Fanelli, Cristiano and Giroux, James, ELUQuant: Event-level Uncertainty Quantification in Deep Inelastic Scattering (2024). Machine Learning: Science and Technology, 5(1).
https://doi.org/10.1088/2632-2153/ad2098
DOI
https://doi.org/10.1088/2632-2153/ad2098
