Journals of Accelerator Conferences Website (JACoW)

JACoW is a publisher in Geneva, Switzerland that publishes the proceedings of accelator conferences held around the world by an international collaboration of editors.

URLhttps://doi.org/10.18429/JACoW-IPAC2024-TUPS60
TitleTowards few-shot reinforcement learning in particle accelerator control
Authors
  • S. Hirlaender, L. Lamminger, S. Pochaba
    University of Salzburg
  • A. Santamaria Garcia, C. Xu, L. Scomparin
    Karlsruhe Institute of Technology
  • J. Kaiser
    Deutsches Elektronen-Synchrotron
  • V. Kain
    European Organization for Nuclear Research
AbstractThis paper addresses the automation of particle accelerator control through reinforcement learning (RL). It highlights the potential to increase reliable performance, especially in light of new diagnostic tools and the increasingly complex variable schedules of specific accelerators. We focus on the physics simulation of the AWAKE electron line, an ideal platform for performing in-depth studies that allow a clear distinction between the problem and the performance of different algorithmic approaches for accurate analysis. The main challenges are the lack of realistic simulations and partially observable environments. We show how effective results can be achieved through meta-reinforcement learning, where an agent is trained to quickly adapt to specific real-world scenarios based on prior training in a simulated environment with variable unknowns. When suitable simulations are lacking or too costly, a model-based method using Gaussian processes is used for direct training in a few shots only. The work opens new avenues for implementing control automation in particle accelerators, significantly increasing their efficiency and adaptability.
Paperdownload: TUPS60.pdf
CiteBibTeX, LaTeX, Text/Word, RIS, EndNote
Conference15th International Particle Accelerator Conference
Series
LocationNashville, TN
Date19-24 May 2024
PublisherJACoW Publishing, Geneva, Switzerland
Editorial BoardFulvia Pilat - Oak Ridge National Laboratory Wolfram Fischer - Brookhaven National Laboratory Robert Saethre - Oak Ridge National Laboratory Petr Anisimov - Los Alamos National Laboratory Ivan Andrian - Elettra-Sincrotrone Trieste S.C.p.A.
Online ISBN978-3-95450-247-9
Online ISSN2673-5490
Received15 May 2024
Revised19 May 2024
Accepted20 May 2024
Issued01 July 2024
DOI10.18429/JACoW-IPAC2024-TUPS60
Pages1804-1807