Bayesian optimization of a free-electron laser

Abstract

The Linac Coherent Light Source changes configurations multiple times per day, necessitating fast tuning strategies to reduce setup time for successive experiments. To this end, we employ a Bayesian approach to transport optics tuning to optimize groups of quadrupole magnets. We use a Gaussian process to provide a probabilistic model of the machine response with respect to control parameters from a modest number of samples. Subsequent samples are selected during optimization using a statistical test combining the model prediction and uncertainty. The model parameters are fit from archived scans, and correlations between devices are added from a simple beam transport model. The result is a sample-efficient optimization routine, which we show significantly outperforms existing optimizers.

6 pages, 3 figures

References in corpus (6)

• Scikit-learn: Machine Learning in Python
• A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning
• Structure Discovery in Nonparametric Regression through Compositional Kernel Search
• Stochastic Variational Deep Kernel Learning
• Differentiable Compositional Kernel Learning for Gaussian Processes
• Online optimization of storage ring nonlinear beam dynamics

Cited by in corpus (47)

• Machine Learning in Nuclear Physics
• Automation and control of laser wakefield accelerators using Bayesian optimisation
• 2022 Review of Data-Driven Plasma Science
• An Ultra-Compact X-Ray Free-Electron Laser
• Multi-Objective Bayesian Optimization for Accelerator Tuning
• Physics-informed Gaussian Process for Online Optimization of Particle Accelerators
• Bayesian optimal control of GHZ states in Rydberg lattices
• Bayesian optimization of laser-plasma accelerators assisted by reduced physical models
• Cheetah: Bridging the Gap Between Machine Learning and Particle Accelerator Physics with High-Speed, Differentiable Simulations
• Differentiable Preisach Modeling for Characterization and Optimization of Accelerator Systems with Hysteresis
• The Continuous Electron Beam Accelerator Facility at 12 GeV
• Tuning Particle Accelerators with Safety Constraints using Bayesian Optimization
• Accurate and confident prediction of electron beam longitudinal properties using spectral virtual diagnostics
• Multi-Objective Bayesian Active Learning for MeV-ultrafast electron diffraction
• Virtual-Diagnostic-Based Time Stamping for Ultrafast Electron Diffraction
• Construction and Commissioning of Mid-Infrared SASE FEL at cERL
• Bayesian Optimization of the Beam Injection Process into a Storage Ring
• Use of Bayesian Optimization to Understand the Structure of Nuclei
• Online Bayesian Optimization for a Recoil Mass Separator
• A data management system for machine learning research of tokamak
• Beam shaping using an ultra-high vacuum multileaf collimator and emittance exchange beamline
• Injection Optimization at Particle Accelerators via Reinforcement Learning: From Simulation to Real-World Application
• Optimization of the injection beam line at the Cooler Synchrotron COSY using Bayesian Optimization
• Neural Process for Black-Box Model Optimization Under Bayesian Framework
• Mixed Diagnostics for Longitudinal Properties of Electron Bunches in a Free-Electron Laser
• Multipoint-BAX: A New Approach for Efficiently Tuning Particle Accelerator Emittance via Virtual Objectives
• Real-time Artificial Intelligence for Accelerator Control: A Study at the Fermilab Booster
• Derivative-Free Optimization of a Rapid-Cycling Synchrotron
• Optimization of transformer ratio and beam loading in a plasma wakefield accelerator with a structure-exploiting algorithm
• Bayesian Optimization Algorithms for Accelerator Physics
• An optimization method to compensate accelerator performance drifts
• Online tuning and light source control using a physics-informed Gaussian process Adi
• Optimisation of the Accelerator Control by Reinforcement Learning: A Simulation-Based Approach
• Automated Discovery of Coupled Mode Setups
• On the development of a practical Bayesian optimisation algorithm for expensive experiments and simulations with changing environmental conditions
• Autonomous Electron Tomography Reconstruction with Machine Learning
• A Hybrid Neural Architecture: Online Attosecond X-ray Characterization
• Time-inversion of spatiotemporal beam dynamics using uncertainty-aware latent evolution reversal
• AI-Assisted Transport of Radioactive Ion Beams
• Developing Robust Digital Twins and Reinforcement Learning for Accelerator Control Systems at the Fermilab Booster
• Optimizing High-Dimensional Physics Simulations via Composite Bayesian Optimization
• Fast Posterior Estimation of Cardiac Electrophysiological Model Parameters via Bayesian Active Learning
• Adaptive Latent Space Tuning for Non-Stationary Distributions
• Bias-Robust Bayesian Optimization via Dueling Bandits
• Adaptive Machine Learning for Time-Varying Systems: Low Dimensional Latent Space Tuning
• Robustness Optimization for Compact Free-electron Laser Driven by Laser Wakefield Accelerators
• Modeling of Advanced Accelerator Concepts