Physics-Informed Neural Networks (PINNs) For DVCS Cross Sections

Presenter and Co-Authors

Manal Almaeen, Old Dominion UniversityFollow
Jake Grigsby, University of Virginia
Joshua Hoskins, University of Virginia
Brandon Kriesten, Center for Nuclear Femtography
Yaohang Li, Old Dominion UniversityFollow
Huey-Wen Lin, Michigan State University
Simonetta Liuti, University of Virginia
Sorawich Maichum, University of Virginia

College

College of Sciences

Department

Computer Science

Graduate Level

Doctoral

Publication Date

4-2022

DOI

10.25883/nndj-pz83

Abstract

We present a physics informed deep learning technique for Deeply Virtual Compton Scattering (DVCS) cross sections from an unpolarized proton target using both an unpolarized and polarized electron beam. Training a deep learning model typically requires a large size of data that might not always be available or possible to obtain. Alternatively, a deep learning model can be trained using additional knowledge gained by enforcing some physics constraints such as angular symmetries for better accuracy and generalization. By incorporating physics knowledge to our deep learning model, our framework shows precise predictions on the DVCS cross sections and better extrapolation on unseen kinematics compared to the basic deep learning approaches.

Keywords

Physics informed deep learning/ DVCS

Disciplines

Artificial Intelligence and Robotics | Computer Sciences | Nuclear | Physics

Files

Recommended Citation

Almaeen, Manal; Grigsby, Jake; Hoskins, Joshua; Kriesten, Brandon; Li, Yaohang; Lin, Huey-Wen; Liuti, Simonetta; and Maichum, Sorawich, "Physics-Informed Neural Networks (PINNs) For DVCS Cross Sections" (2022). College of Sciences Posters. 14.
https://digitalcommons.odu.edu/gradposters2022_sciences/14