SYRIPY: Differentiable Synchrotron Radiation Inference for Beam Diagnostics

What is it about?

The article presents SYRIPY, a software package for performing inference of synchrotron-radiation-based diagnostics. SYRIPY uses PyTorch, making it both differentiable and able to leverage the high performance of GPUs. The package consists of three modules: a particle tracker, Lienard-Wiechert solver, and Fourier optics propagator, allowing start-to-end simulation of synchrotron radiation detection. The article also includes a description of the theory and numerical implementation used to simulate synchrotron radiation production and detection, as well as benchmark simulations and a section on gradient-based Bayesian inference.

Featured Image

Why is it important?

The research presented in this text is important for several reasons: It addresses the challenge of traditional diagnostics for high-intensity particle accelerators, which pose a risk of destroying the diagnostic equipment itself. It proposes a non-invasive, single-shot diagnostic method using synchrotron radiation, which can be used in conjunction with experiments. It introduces a new software package, SYRIPY, which can perform Bayesian inference of beam parameters using stochastic variational inference. SYRIPY is based on PyTorch, which makes it both differentiable and able to leverage the high performance of GPUs, two vital characteristics for performing statistical inference. The development of non-invasive diagnostics for high-intensity particle accelerators is crucial for the advancement of research in various fields, including plasma wakefield accelerators and strong field quantum electrodynamics. The ability to perform statistical inference using Bayesian methods allows for more accurate and efficient analysis of beam parameters, further enhancing the capabilities of particle accelerators. Key Takeaways: 1. The research presents a non-invasive diagnostic method using synchrotron radiation, which is a promising alternative to traditional diagnostics for high-intensity particle accelerators. 2. SYRIPY is a new software package that uses stochastic variational inference to perform Bayesian inference of beam parameters, and it is based on PyTorch, which makes it both differentiable and able to leverage the high performance of GPUs. 3. The research demonstrates that SYRIPY can accurately infer beam parameters using stochastic variational inference, and it has been benchmarked against SRW, the prevalent numerical package in the field, showing good agreement and up to a 50% speed improvement.