Denoising diffusion models with geometry adaptation for high fidelity calorimeter simulation

TL;DR

CaloDiffusion is a novel 3D diffusion model with geometry adaptation that efficiently generates high-fidelity calorimeter simulations, closely matching full physics-based simulations for collider data analysis.

Contribution

The paper introduces CaloDiffusion, a diffusion model with geometry-aware layers and a new GLaM component, enabling accurate calorimeter shower generation across irregular detector geometries.

Findings

Generated showers are nearly indistinguishable from full simulations.

The model effectively handles irregular geometries with GLaM.

High fidelity results demonstrate potential for collider data analysis.

Abstract

Simulation is crucial for all aspects of collider data analysis, but the available computing budget in the High Luminosity LHC era will be severely constrained. Generative machine learning models may act as surrogates to replace physics-based full simulation of particle detectors, and diffusion models have recently emerged as the state of the art for other generative tasks. We introduce CaloDiffusion, a denoising diffusion model trained on the public CaloChallenge datasets to generate calorimeter showers. Our algorithm employs 3D cylindrical convolutions, which take advantage of symmetries of the underlying data representation. To handle irregular detector geometries, we augment the diffusion model with a new geometry latent mapping (GLaM) layer to learn forward and reverse transformations to a regular geometry that is suitable for cylindrical convolutions. The showers generated by our approach are nearly indistinguishable from the full simulation, as measured by several different metrics.