Cheetah - Bridging the Gap Between Machine Learning and Particle Accelerator Physics with High-Speed, Differentiable Simulations

Aug 3, 2026

Upload by Riley

Machine learning is increasingly used to address modern challenges in accelerator physics, yet beam-time scarcity, costly simulations, and high-dimensional optimisation restrict the training data available for state-of-the-art models. This work presents Cheetah, a PyTorch-based high-speed differentiable linear-beam dynamics code that accelerates simulation and enables efficient gradient-based optimisation for accelerator tuning and system identification. Five demonstrations cover reinforcement learning, gradient-based beamline tuning, gradient-based system identification, physics-informed Bayesian optimisation priors, and modular neural surrogate modelling of space charge effects.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages16

Reading Time40 min

AccessFree to download

Cheetah: Bridging the Gap Between Machine Learning and Particle Accelerator Physics

Introduction

Data bottlenecks in machine learning for accelerators

High-speed differentiable simulation with Cheetah

Applications and examples

arXiv :2401 .05815v1 [physics .acc-ph] 11 Jan 2024 Cheetah: Bridging the Gap Between Machine Learning and Particle Accelerator Physics with High-Speed, Differentiable Simulations ∗ Jan Kaiser, 1,† Chenran Xu,2,‡ Annika Eichler, 1, 3 and Andrea Santamaria Garcia2 1 Deutsches Elektronen-Synchrotron DESY, Germany 2 Karlsruhe Institute of Technology (KIT), Germany 3 Hamburg University of Technology, 21073 Hamburg, Germany (Dated: 11 January 2024) Machine learning has emerged as a powerful solution to the modern challenges in accelerator physics. However, the limited availability of beam time, the computational cost of simulations, and the high-dimensionality of optimisation problems pose significant challenges in generating the required data for training state-of-the-art machine learning models. In this work, we introduce Cheetah, a PyTorch-based high-speed differentiable linear-beam dynamics code. Cheetah enables the fast collection of large data sets by reducing computation times by multiple orders of magnitude and facilitates efficient gradient-based optimisation for accelerator tuning and system identification. This positions Cheetah as a user-friendly, readily extensible tool that integrates seamlessly with widely adopted machine learning tools. We showcase the utility of Cheetah through five examples, including reinforcement learning training, gradient-based beamline tuning, gradient-based system identification, physics-informed Bayesian optimisation priors, and modular neural network surrogate modelling of space charge effects. The use of such a high-speed differentiable simulation code will simplify the development of machine learning-based methods for particle accelerators and fast-track their integration into everyday operations of accelerator facilities. I. INTRODUCTION Future particle accelerator experiments will place everincreasing demands on the performance and capabilities of particle accelerator operations and experiment analysis. In order to meet these demands, the research community is increasingly turning to machine learning (ML) methods, which have already demonstrated their ability to push the envelope of what is possible in the field of accelerator science [1–4] . One of the remaining challenges holding back this line of research is the demand for large amounts of data (including environment interactions) of these methods. Reinforcement learning (RL), for example, has already successfully been used to train intelligent tuning algorithms and controllers that can outperform the currently deployed black-box optimisation algorithms and handcrafted controllers [1, 5–7] . However, RL methods require many interactions with their target task to train a well-performing policy. For example, 6000000 samples were needed in [1] to successfully train a policy on a transverse beam tuning task. Orders of magnitude larger number of samples are also common with other RL applications [8, 9] . The general scarcity of beam time makes collecting experimental data for ML methods, such as RL, a significant bottleneck. Gathering at least partial data sets in simulation can alleviate this problem, but existing accelerator simulation codes have mostly been ∗ All figures and pictures by the authors are published under a CC-BY7 licence. † jan.kaiser@desy.de; Equal contributions ‡ [chenran.xu@kit.edu](chenran.xu@kit.edu); Equal contributions developed with a focus on the design phase of accelerators, where high-fidelity and physical correctness are critical, and computing times range from minutes to several hours for one simulation. Consequently, data collection with existing simulation codes becomes impractical with the growing demand for large data sets. In this paper, we introduce Cheetah, a PyTorch-based high-speed differentiable linear-beam dynamics code. Cheetah is capable of accelerating beam dynamics simulations by multiple orders of magnitude through tensorised computation and several speed optimisati

FAQs

What problem does Cheetah address for machine learning in accelerator physics?

It addresses the bottlenecks of limited beam time, expensive simulation cost, and the large data needs of ML methods by providing a fast differentiable simulation tool for generating training data and supporting optimisation.

How does Cheetah speed up beam dynamics simulations?

Cheetah accelerates simulations by multiple orders of magnitude using tensorised computation and additional speed optimisation methods, while being implemented in PyTorch to support automatic differentiation.

What types of tasks are demonstrated using Cheetah?

The document presents five examples: reinforcement learning training, gradient-based beamline tuning, gradient-based system identification, physics-informed Bayesian optimisation priors, and differentiable neural surrogate modelling of space-charge effects.

Aug 3, 2026

Upload by Riley