Cross-Geometry Transfer Learning in Fast Electromagnetic Shower Simulation

TL;DR

This paper introduces a transfer learning framework for fast, geometry-adaptive electromagnetic shower simulation using point cloud models, significantly reducing data needs and enabling efficient adaptation across detector designs.

Contribution

It presents a novel transfer learning approach for calorimeter simulation that adapts to different geometries without full retraining, improving efficiency and flexibility.

Findings

44% improvement in Wasserstein distance with only 100 target samples

Parameter-efficient fine-tuning updates only 17% of model parameters

Framework handles new geometries without re-voxelizing showers

Abstract

Accurate particle shower simulation remains a critical computational bottleneck for high-energy physics. Traditional Monte Carlo methods, such as Geant4, are computationally prohibitive, while existing machine learning surrogates are tied to specific detector geometries and require complete retraining for each design change or alternative detector. We present a transfer learning framework for generative calorimeter simulation models that enables adaptation across diverse geometries with high data efficiency. Using point cloud representations and pre-training on the International Large Detector detector, our approach handles new configurations without re-voxelizing showers for each geometry. On the CaloChallenge dataset, transfer learning with only 100 target-domain samples achieves a $44\%$ improvement on the geometric mean of Wasserstein distance over training from scratch. Parameter-efficient fine-tuning with bias-only adaptation achieves competitive performance while updating only $17\%$ of model parameters. Our analysis provides insight into adaptation mechanisms for particle shower development, establishing a baseline for future progress of point cloud approaches in calorimeter simulation.