15–18 Dec 2026
Duke University
America/New_York timezone
Artificial Intelligence for the Electron Ion Collider (2026)

AI-ready detector data and GNN-based reconstruction for ePIC Barrel Imaging Calorimeter

Not scheduled
15m
LSRC B101 Love Auditorium (Duke University)

LSRC B101 Love Auditorium

Duke University

Durham, NC 27710
AI/ML for ePIC and Beyond AI/ML for ePIC and Beyond

Speaker

Zhiwan Xu (Argonne National Lab)

Description

Next-generation nuclear physics experiments at the Electron-Ion Collider (EIC) require advanced reconstruction and analysis techniques capable of exploiting the increasingly detailed data provided by the ePIC detector systems. In this work, we present the design of an AI-ready data framework based on low-level pre-reconstruction detector readout for general detectors, and discuss its application to the Barrel Imaging electromagnetic Calorimeter (BIC) of the ePIC detector at the EIC. The BIC hybrid readout combines high-spatial-resolution AstroPix layers with silicon sensors and Pb/scintillating-fiber layers for energy sampling, demonstrating an ideal benchmark with multi-modality dataset for the AI study. We also present and discuss a Graph Neural Network (GNN)-based approach for end-to-end AI reconstruction using the new AI-ready data, including shower clustering, particle reconstructing and classification.

Authors

Maria Zurek (Argonne National Laboratory) Sylvester Joosten (Argonne National Laboratory) Zhiwan Xu (Argonne National Lab)

Presentation materials

There are no materials yet.