Title |
Bayesian Optimisation for Fast and Safe Parameter Tuning of SwissFEL |
Authors |
- J. Kirschner, A. Krause, M. Mutný, M. Nonnenmacher
ETH, Zurich, Switzerland
- A. Adelmann, N. Hillerpresenter, R. Ischebeck
PSI, Villigen PSI, Switzerland
|
Abstract |
Parameter tuning is a notoriously time-consuming task in accelerator facilities. As tool for global optimization with noisy evaluations, Bayesian optimization was recently shown to outperform alternative methods. By learning a model of the underlying function using all available data, the next evaluation can be chosen carefully to find the optimum with as few steps as possible and without violating any safety constraints. However, the per-step computation time increases significantly with the number of parameters and the generality of the approach can lead to slow convergence on functions that are easier to optimize. To overcome these limitations, we divide the global problem into sequential subproblems that can be solved efficiently using safe Bayesian optimization. This allows us to trade off local and global convergence and to adapt to additional structure in the objective function. Further, we provide slice-plots of the function as user feedback during the optimization. We showcase how we use our algorithm to tune up the FEL output of SwissFEL with up to 40 parameters simultaneously, and reach convergence within reasonable tuning times in the order of 30 minutes (< 2000 steps).
|
Paper |
download THP061.PDF [0.233 MB / 4 pages] |
Export |
download ※ BibTeX
※ LaTeX
※ Text/Word
※ RIS
※ EndNote |
Conference |
FEL2019 |
Series |
Free Electron Laser Conference (39th) |
Location |
Hamburg, Germany |
Date |
26-30 August 2019 |
Publisher |
JACoW Publishing, Geneva, Switzerland |
Editorial Board |
Volker R.W. Schaa (GSI, Darmstadt, Germany); Winfried Decking (DESY, Hamburg, Germany); Harald Sinn (EuXFEL, Schenefeld, Germany); Gianluca Geloni (EuXFEL, Schenefeld, Germany); Siegfried Schreiber (DESY, Hamburg, Germany); Michaela Marx (DESY, Hamburg, Germany) |
Online ISBN |
978-3-95450-210-3 |
Online ISSN |
"" |
Received |
13 August 2019 |
Accepted |
27 August 2019 |
Issue Date |
05 November 2019 |
DOI |
doi:10.18429/JACoW-FEL2019-THP061 |
Pages |
707-710 |
Copyright |
Published by JACoW Publishing under the terms of the Creative Commons Attribution 3.0 International license. Any further distribution of this work must maintain attribution to the author(s), the published article's title, publisher, and DOI. |
|