Speaker
Mr
Yu-Sheng Liu
(National Kaohsiung Normal University)
Description
We present an application of generative AI to improve shower profile modeling in grid-style detectors, with a focus on the CsI calorimeter used in the KOTO experiment. The core idea is to employ generative models to reconstruct electromagnetic shower patterns at a resolution finer than the detector’s actual segmentation. This approach enables the generation of high-resolution profiles that recover spatial features often lost in measurements. These enhanced profiles have the potential to improve analyses that rely on shower shape information, such as the reconstruction of the incident particle’s direction and particle identification.
Author
Mr
Yu-Sheng Liu
(National Kaohsiung Normal University)
Co-author
Yu-Chen Tung
(National Kaohsiung Normal University)