Geometry-aware Autoregressive Models for Calorimeter Shower Simulations
This addresses the problem of high simulation time and poor generalization for particle physics researchers, offering a potential replacement for hundreds of models in experiments like the Large Hadron Collider, though it is incremental as it builds on existing generative methods.
The paper tackles the bottleneck of calorimeter shower simulations in particle physics by developing a geometry-aware autoregressive model that adapts energy deposition based on cell size and position, achieving a proof-of-concept for generalization to new geometries with minimal training.
Calorimeter shower simulations are often the bottleneck in simulation time for particle physics detectors. A lot of effort is currently spent on optimizing generative architectures for specific detector geometries, which generalize poorly. We develop a geometry-aware autoregressive model on a range of calorimeter geometries such that the model learns to adapt its energy deposition depending on the size and position of the cells. This is a key proof-of-concept step towards building a model that can generalize to new unseen calorimeter geometries with little to no additional training. Such a model can replace the hundreds of generative models used for calorimeter simulation in a Large Hadron Collider experiment. For the study of future detectors, such a model will dramatically reduce the large upfront investment usually needed to generate simulations.