Refinement of AI-assisted Track Reconstruction and Particle Identification for CLAS12 ALERT Experiment at Jefferson Lab
ORAL
Abstract
Understanding the in-medium modifications of bound nucleon structure remains a fundamental challenge in hadronic physics. The deeply virtual Compton scattering (DVCS) process offers a promising approach to probe the partonic structure of light nuclei, providing access to their three-dimensional distributions through Generalized Parton Distributions (GPDs) and thus studying the associated in-medium stimulated effects.
The recently conducted CLAS12 experiment at Jefferson Lab employed the newly built A Low Energy Recoil Tracker (ALERT) detector to study tagged DVCS by impinging an 11~GeV polarized electron beam on $^4$He or $^2$H. The ALERT detector is comprised of a hyperbolic drift chamber (AHDC) and a time-of-flight (ATOF) array to provide an effective separation of low-momentum nuclear-recoil fragments, including $^1$H, $^2$H, $^3$H, $^3$He, and $^4$He, down to 70~MeV/c across a broad kinematic range.
Recent advances in artificial intelligence (AI), including new model architectures, have proven effective for high-rate experiments with substantially elevated background noise, such as AHDC in ALERT. AI is deployed in ALERT experiments to enhance AHDC track-finding efficiency, purity, and speed compared to conventional algorithms, as well as to identify the recoil fragments. In this talk, an overview of the ALERT physics program will be provided, along with an update on the ongoing development and optimization of the AI-assisted track reconstruction and particle identification for simulation and real data.
The recently conducted CLAS12 experiment at Jefferson Lab employed the newly built A Low Energy Recoil Tracker (ALERT) detector to study tagged DVCS by impinging an 11~GeV polarized electron beam on $^4$He or $^2$H. The ALERT detector is comprised of a hyperbolic drift chamber (AHDC) and a time-of-flight (ATOF) array to provide an effective separation of low-momentum nuclear-recoil fragments, including $^1$H, $^2$H, $^3$H, $^3$He, and $^4$He, down to 70~MeV/c across a broad kinematic range.
Recent advances in artificial intelligence (AI), including new model architectures, have proven effective for high-rate experiments with substantially elevated background noise, such as AHDC in ALERT. AI is deployed in ALERT experiments to enhance AHDC track-finding efficiency, purity, and speed compared to conventional algorithms, as well as to identify the recoil fragments. In this talk, an overview of the ALERT physics program will be provided, along with an update on the ongoing development and optimization of the AI-assisted track reconstruction and particle identification for simulation and real data.
–
Presenters
-
Mathieu Ouillon
Mississippi State University
Authors
-
Mathieu Ouillon
Mississippi State University