Speaker
Description
Generative machine learning models offer a promising way to simulate events. Given the already high computational cost of simulation and the expected increase in data in the high-precision era of the LHC and at future colliders, such fast surrogate simulators are urgently needed.
This contribution presents initial progress towards accurately simulating of hadronic showers in a highly granular scintillator calorimeter for future colliders. We used two generative models in this study: a Wasserstein-GAN (WGAN) and Bounded Information Bottleneck Autoencoder (BIB-AE). Then we compare the achieved simulation quality before and after interfacing with the state-of-the-art pattern recognition algorithm used by ILD, the so-called PandoraPFA. This brings generative models one step closer to practical applications.
| 1st preferred time slot for your oral presentation | 10:00-12:00 JST (3:00-5:00 CEST, 21:00-23:00 EDT, 18:00-20:00 PDT) |
|---|---|
| 2nd preferred time slot for your oral presentation | 15:30-17:30 JST (8:30-10:30 CEST, 2:30-4:30 EDT, 23:30-1:30 PDT) |
