Bayesian Optimization Algorithms for Accelerator Physics

Accelerator physics relies on numerical algorithms to solve optimization problems in online accelerator control and tasks such as experimental design and model calibration in simulations. The effectiveness of optimization algorithms in discovering ideal solutions for complex challenges with limited resources often determines the problem complexity these methods can address. The accelerator physics community has recognized the advantages of Bayesian optimization algorithms, which leverage statistical surrogate models of objective functions to effectively address complex optimization challenges, especially in the presence of noise during accelerator operation and in resource-intensive physics simulations. In this review article, we offer a conceptual overview of applying Bayesian optimization techniques towards solving optimization problems in accelerator physics. We begin by providing a straightforward explanation of the essential components that make up Bayesian optimization techniques. We then give an overview of current and previous work applying and modifying these techniques to solve accelerator physics challenges. Finally, we explore practical implementation strategies for Bayesian optimization algorithms to maximize their performance, enabling users to effectively address complex optimization challenges in real-time beam control and accelerator design.

[1]  E. Bakshy,et al.  Unexpected Improvements to Expected Improvement for Bayesian Optimization , 2023, Neural Information Processing Systems.

[2]  R. Roussel,et al.  Demonstration of Autonomous Emittance Characterization at the Argonne Wakefield Accelerator , 2023, Instruments.

[3]  W. Leemans,et al.  Tuning curves for a laser-plasma accelerator , 2023, Physical Review Accelerators and Beams.

[4]  H. Schlarb,et al.  Learning to Do or Learning While Doing: Reinforcement Learning and Bayesian Optimisation for Online Continuous Tuning , 2023, ArXiv.

[5]  J. Susini,et al.  The Extremely Brilliant Source storage ring of the European Synchrotron Radiation Facility , 2023, Communications physics.

[6]  J. Vay,et al.  Bayesian optimization of laser-plasma accelerators assisted by reduced physical models , 2022, Physical Review Accelerators and Beams.

[7]  M. Schuh,et al.  Bayesian optimization of the beam injection process into a storage ring , 2022, Physical Review Accelerators and Beams.

[8]  R. Roussel,et al.  Neural Network Prior Mean for Particle Accelerator Injector Tuning , 2022, 2211.09028.

[9]  J. Zhang,et al.  Fast Optimization for Betatron Radiation from Laser Wakefield Acceleration Based on Bayesian Optimization , 2022, SSRN Electronic Journal.

[10]  S. Karsch,et al.  Multi-objective and multi-fidelity Bayesian optimization of laser-plasma acceleration , 2022, Physical Review Research.

[11]  W. Neiswanger,et al.  Multipoint-BAX: a new approach for efficiently tuning particle accelerator emittance via virtual objectives , 2022, Mach. Learn. Sci. Technol..

[12]  D. Ratner,et al.  Phase Space Reconstruction from Accelerator Beam Measurements Using Neural Networks and Differentiable Simulations. , 2022, Physical Review Letters.

[13]  A. Scheinker,et al.  Extremum Seeking-Based Control System for Particle Accelerator Beam Loss Minimization , 2022, IEEE Transactions on Control Systems Technology.

[14]  Xiaobiao Huang,et al.  Optimization method to compensate accelerator performance drifts , 2022, Physical Review Accelerators and Beams.

[15]  A. Krause,et al.  Tuning Particle Accelerators with Safety Constraints using Bayesian Optimization , 2022, Physical Review Accelerators and Beams.

[16]  J. Donini,et al.  Toward the end-to-end optimization of particle physics instruments with differentiable programming , 2022, Reviews in Physics.

[17]  G. Hartmann,et al.  Optimizing a superconducting radio-frequency gun using deep reinforcement learning , 2022, Physical Review Accelerators and Beams.

[18]  R. Roussel,et al.  Differentiable Preisach Modeling for Characterization and Optimization of Particle Accelerator Systems with Hysteresis. , 2022, Physical Review Letters.

[19]  M. S. Smith,et al.  Online Bayesian optimization for a recoil mass separator , 2022, Physical Review Accelerators and Beams.

[20]  K. Brown,et al.  Bayesian optimization experiment for trajectory alignment at the low energy RHIC electron cooling system , 2022, Physical Review Accelerators and Beams.

[21]  A. Döpp,et al.  Leveraging Trust for Joint Multi-Objective and Multi-Fidelity Optimization , 2021, Machine Learning: Science and Technology.

[22]  Louis Emery,et al.  Application of a machine learning based algorithm to online optimization of the nonlinear beam dynamics of the Argonne Advanced Photon Source , 2021, Physical Review Accelerators and Beams.

[23]  Michael Borland,et al.  Experience with On-line Optimizers for APS Linac Front End Optimization , 2021 .

[24]  Aaron Stein,et al.  Gaussian processes for autonomous data acquisition at large-scale synchrotron and neutron facilities , 2021, Nature Reviews Physics.

[25]  Young-Kee Kim,et al.  Turn-key constrained parameter space exploration for particle accelerators using Bayesian active learning , 2021, Nature Communications.

[26]  D. Ratner,et al.  Recovering the phase and amplitude of X-ray FEL pulses using neural networks and differentiable models. , 2021, Optics express.

[27]  Ke Alexander Wang,et al.  Bayesian Algorithm Execution: Estimating Computable Properties of Black-box Functions Using Mutual Information , 2021, ICML.

[28]  Stefan M. Wild,et al.  libEnsemble: A Library to Coordinate the Concurrent Evaluation of Dynamic Ensembles of Calculations , 2021, IEEE Transactions on Parallel and Distributed Systems.

[29]  R. Lehe,et al.  Bayesian Optimization of a Laser-Plasma Accelerator. , 2021, Physical review letters.

[30]  Martin Jankowiak,et al.  High-Dimensional Bayesian Optimization with Sparse Axis-Aligned Subspaces , 2021, UAI.

[31]  R. Allmendinger,et al.  Safe Learning and Optimization Techniques: Towards a Survey of the State of the Art , 2021, TAILOR.

[32]  Gianluca Valentino,et al.  Sample-efficient reinforcement learning for CERN accelerator control , 2020, Physical Review Accelerators and Beams.

[33]  N. Bourgeois,et al.  Automation and control of laser wakefield accelerators using Bayesian optimization , 2020, Nature Communications.

[34]  Zhe Zhang,et al.  Online accelerator optimization with a machine learning-based stochastic algorithm , 2020, Mach. Learn. Sci. Technol..

[35]  A. Hanuka,et al.  Multiobjective Bayesian optimization for online accelerator tuning , 2020, Physical Review Accelerators and Beams.

[36]  Xiaobiao Huang,et al.  Storage ring nonlinear dynamics optimization with multi-objective multi-generation Gaussian process optimizer , 2020 .

[37]  Adi Hanuka,et al.  Physics-informed Gaussian Process for Online Optimization of Particle Accelerators , 2020, Physical Review Accelerators and Beams.

[38]  Jason K. Streit,et al.  Autonomous materials discovery driven by Gaussian process regression with inhomogeneous measurement noise and anisotropic kernels , 2020, Scientific Reports.

[39]  Maximilian Balandat,et al.  Differentiable Expected Hypervolume Improvement for Parallel Multi-Objective Bayesian Optimization , 2020, NeurIPS.

[40]  Alexander Scheinker,et al.  Online multi-objective particle accelerator optimization of the AWAKE electron beam line for simultaneous emittance and orbit control , 2020, 2003.11155.

[41]  Matthias Poloczek,et al.  Scalable Constrained Bayesian Optimization , 2020, AISTATS.

[42]  Natalia Gimelshein,et al.  PyTorch: An Imperative Style, High-Performance Deep Learning Library , 2019, NeurIPS.

[43]  C. Melton,et al.  Demonstration of Machine Learning-Based Model-Independent Stabilization of Source Properties in Synchrotron Light Sources. , 2019, Physical review letters.

[44]  S. Reimann,et al.  Beam Line Optimization using Derivative-Free Algorithms , 2019, Journal of Physics: Conference Series.

[45]  Felix Berkenkamp,et al.  Safe Exploration for Interactive Machine Learning , 2019, NeurIPS.

[46]  Matthias Poloczek,et al.  Scalable Global Optimization via Local Bayesian Optimization , 2019, NeurIPS.

[47]  S. Ermon,et al.  Bayesian Optimization of a Free-Electron Laser. , 2019, Physical review letters.

[48]  Ruipeng Li,et al.  A Kriging-Based Approach to Autonomous Experimentation with Applications to X-Ray Scattering , 2019, Scientific Reports.

[49]  Xiaobiao Huang,et al.  Multi-objective multi-generation Gaussian process optimizer for design optimization , 2019, ArXiv.

[50]  Joseph R. Kasprzyk,et al.  Introductory overview: Optimization using evolutionary algorithms and other metaheuristics , 2019, Environ. Model. Softw..

[51]  Eytan Bakshy,et al.  Bayesian Optimization for Policy Search via Online-Offline Experimentation , 2019, J. Mach. Learn. Res..

[52]  A. Adelmann,et al.  Machine learning for orders of magnitude speedup in multiobjective optimization of particle accelerator systems , 2019, Physical Review Accelerators and Beams.

[53]  Andreas Krause,et al.  Adaptive and Safe Bayesian Optimization in High Dimensions via One-Dimensional Subspaces , 2019, ICML.

[54]  Xiaobiao Huang,et al.  Robust simplex algorithm for online optimization , 2018, Physical Review Accelerators and Beams.

[55]  Andrew Gordon Wilson,et al.  GPyTorch: Blackbox Matrix-Matrix Gaussian Process Inference with GPU Acceleration , 2018, NeurIPS.

[56]  Alexander Scheinker,et al.  Demonstration of Model-Independent Control of the Longitudinal Phase Space of Electron Beams in the Linac-Coherent Light Source with Femtosecond Resolution. , 2018, Physical review letters.

[57]  Joel W. Burdick,et al.  Stagewise Safe Bayesian Optimization with Gaussian Processes , 2018, ICML.

[58]  Frank Hutter,et al.  Maximizing acquisition functions for Bayesian optimization , 2018, NeurIPS.

[59]  Stephen J. Roberts,et al.  Bayesian Optimization for Dynamic Problems , 2018, 1803.03432.

[60]  Peter I. Frazier,et al.  Continuous-fidelity Bayesian Optimization with Knowledge Gradient , 2018 .

[61]  Frank Hutter,et al.  The reparameterization trick for acquisition functions , 2017, ArXiv.

[62]  Sandra Biedron,et al.  Using a Neural Network Control Policy for Rapid Switching Between Beam Parameters in an FEL , 2017 .

[63]  Jeffrey Larson,et al.  Photoinjector optimization using a derivative-free, model-based trust-region algorithm for the Argonne Wakefield Accelerator , 2017, Journal of Physics: Conference Series.

[64]  Kirthevasan Kandasamy,et al.  Multi-fidelity Bayesian Optimisation with Continuous Approximations , 2017, ICML.

[65]  Aaron Klein,et al.  Bayesian Optimization with Robust Bayesian Neural Networks , 2016, NIPS.

[66]  Andrew Gordon Wilson,et al.  Stochastic Variational Deep Kernel Learning , 2016, NIPS.

[67]  Sebastian Ruder,et al.  An overview of gradient descent optimization algorithms , 2016, Vestnik komp'iuternykh i informatsionnykh tekhnologii.

[68]  Stefano Ermon,et al.  Bayesian Optimization of FEL Performance at LCLS , 2016 .

[69]  Andreas Krause,et al.  Bayesian optimization with safety constraints: safe and automatic parameter tuning in robotics , 2016, Machine Learning.

[70]  Andrew Gordon Wilson,et al.  Deep Kernel Learning , 2015, AISTATS.

[71]  Alkis Gotovos,et al.  Safe Exploration for Optimization with Gaussian Processes , 2015, ICML.

[72]  Neil D. Lawrence,et al.  Batch Bayesian Optimization via Local Penalization , 2015, AISTATS.

[73]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[74]  Matt J. Kusner,et al.  Bayesian Optimization with Inequality Constraints , 2014, ICML.

[75]  Daan Wierstra,et al.  Stochastic Backpropagation and Approximate Inference in Deep Generative Models , 2014, ICML.

[76]  Konstantinos Liagkouras,et al.  An Elitist Polynomial Mutation Operator for Improved Performance of MOEAs in Computer Networks , 2013, 2013 22nd International Conference on Computer Communication and Networks (ICCCN).

[77]  Juhao Wu,et al.  An algorithm for online optimization of accelerators , 2013 .

[78]  Alexander Scheinker,et al.  Model-independent particle accelerator tuning , 2013 .

[79]  J. Maxson,et al.  Demonstration of Low Emittance in the Cornell Energy Recovery Linac Injector Prototype , 2013, 1304.2708.

[80]  Andrew Gordon Wilson,et al.  Gaussian Process Kernels for Pattern Discovery and Extrapolation , 2013, ICML.

[81]  Chong Wang,et al.  Stochastic variational inference , 2012, J. Mach. Learn. Res..

[82]  Andreas Krause,et al.  Contextual Gaussian Process Bandit Optimization , 2011, NIPS.

[83]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[84]  Nando de Freitas,et al.  A Tutorial on Bayesian Optimization of Expensive Cost Functions, with Application to Active User Modeling and Hierarchical Reinforcement Learning , 2010, ArXiv.

[85]  Andrew Gordon Wilson,et al.  Copula Processes , 2010, NIPS.

[86]  S. Ounpraseuth,et al.  Gaussian Processes for Machine Learning , 2008 .

[87]  Edwin V. Bonilla,et al.  Multi-task Gaussian Process Prediction , 2007, NIPS.

[88]  P. Mora.  Laser driven ion acceleration , 2007 .

[89]  Stephen P. Boyd,et al.  Convex Optimization , 2010, IEEE Transactions on Automatic Control.

[90]  M. Borland,et al.  A Parallel Simplex Optimizer and its Application to High-Brightness Storage Ring Design , 2005, Proceedings of the 2005 Particle Accelerator Conference.

[91]  Charles Sinclair,et al.  Multivariate optimization of a high brightness dc gun photoinjector , 2005 .

[92]  L. Emery,et al.  Use of a general-purpose optimization module in accelerator control , 2003, Proceedings of the 2003 Particle Accelerator Conference.

[93]  Kalyanmoy Deb,et al.  A fast and elitist multiobjective genetic algorithm: NSGA-II , 2002, IEEE Trans. Evol. Comput..

[94]  Marc G. Genton,et al.  Classes of Kernels for Machine Learning: A Statistics Perspective , 2002, J. Mach. Learn. Res..

[95]  Gene H. Golub,et al.  Tikhonov Regularization and Total Least Squares , 1999, SIAM J. Matrix Anal. Appl..

[96]  R. Horst,et al.  Global Optimization: Deterministic Approaches , 1992 .

[97]  Charles J. Geyer,et al.  Practical Markov Chain Monte Carlo , 1992 .

[98]  Jorge Nocedal,et al.  On the limited memory BFGS method for large scale optimization , 1989, Math. Program..

[99]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[100]  R. Howard.  Dynamic Programming and Markov Processes , 1960 .

[101]  Academic Programs,et al.  Geneva , 1956, Dostoevsky.

[102]  F. Preisach.  Über die magnetische Nachwirkung , 1935 .

[103]  N. Tran,et al.  Developing Robust Digital Twins and Reinforcement Learning for Accelerator Control Systems at the Fermilab Booster , 1900, Developing Robust Digital Twins and Reinforcement Learning for Accelerator Control Systems at the Fermilab Booster.

[104]  Maximilian Schütte,et al.  A Safe Bayesian Optimization Algorithm for Tuning the Optical Synchronization System at European XFEL , 2023, IFAC-PapersOnLine.

[105]  E. Bründermann,et al.  OPTIMIZATION STUDIES OF SIMULATED TH z RADIATION AT FLUTE , 2022 .

[106]  J. Shtalenkova,et al.  BADGER: THE MISSING OPTIMIZER IN ACR ∗ , 2022 .

[107]  S. Ermon,et al.  Bayesian Algorithm Execution for Tuning Particle Accelerator Emittance with Partial Measurements , 2022, ArXiv.

[108]  R. Roussel.  Proximal Biasing for Bayesian Optimization and Characterization of Physical Systems , 2021 .

[109]  Appendix to: BOTORCH: A Framework for Efficient Monte-Carlo Bayesian Optimization , 2021 .

[110]  Tamim Asfour,et al.  Feedback Design for Control of the Micro-Bunching Instability based on Reinforcement Learning , 2019 .

[111]  J. Qiang.  START-TO-END BEAM DYNAMICS OPTIMIZATION OF X-RAY FEL LIGHT SOURCE ACCELERATORS* , 2017 .

[112]  M. Scholz,et al.  On-Line Optimization of European XFEL with OCELOT , 2017 .

[113]  Nando de Freitas,et al.  Taking the Human Out of the Loop: A Review of Bayesian Optimization , 2016, Proceedings of the IEEE.

[114]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

[115]  Joaquim R. R. A. Martins,et al.  Multimission Aircraft Fuel-Burn Minimization via Multipoint Aerostructural Optimization , 2015 .

[116]  D. Ginsbourger,et al.  Kriging is well-suited to parallelize optimization , 2010 .

[117]  M. Powell.  The BOBYQA algorithm for bound constrained optimization without derivatives , 2009 .

[118]  G. Evans,et al.  Learning to Optimize , 2008 .

[119]  Eduardo F. Camacho,et al.  Introduction to Model Predictive Control , 2007 .

[120]  Kalyanmoy Deb,et al.  Simulated Binary Crossover for Continuous Search Space , 1995, Complex Syst..

[121]  J. Mockus.  The Bayesian Approach to Local Optimization , 1989 .

[122]  John A. Nelder,et al.  A Simplex Method for Function Minimization , 1965, Comput. J..