Accelerator Control using Gaussian Process Model Predictive Control Based Reinforcement Learning

<!--HTML-->Abstract: This project aims to develop a model-based Reinforcement Learning (RL) algorithm, based on the GP-MPC(Gaussian Process-Model Predictive Control) approach, for controlling accelerator systems using Machine Learning. By leveraging insufficient data and uncertainty quantifica...

Descripción completa

Detalles Bibliográficos
Autor principal: Aye, Su
Lenguaje:eng
Publicado: 2023
Materias:
CERN openlab Summer Student Programme 2023
Acceso en línea:http://cds.cern.ch/record/2868378
_version_ 1780978218365353984
author Aye, Su
author_facet Aye, Su
author_sort Aye, Su
collection CERN
description <!--HTML-->Abstract: This project aims to develop a model-based Reinforcement Learning (RL) algorithm, based on the GP-MPC(Gaussian Process-Model Predictive Control) approach, for controlling accelerator systems using Machine Learning. By leveraging insufficient data and uncertainty quantification capabilities of Gaussian Processes, the algorithm will overcome limitations related to beam instrumentation and training large iterations. The final outcome will be the integration of the modified algorithm into CERN's Generic Optimisation Framework, facilitating easy deployment for operational scenarios and enabling intelligent accelerator control.
id cern-2868378
institution Organización Europea para la Investigación Nuclear
language eng
publishDate 2023
record_format invenio
spelling cern-28683782023-08-24T18:50:15Zhttp://cds.cern.ch/record/2868378engAye, SuAccelerator Control using Gaussian Process Model Predictive Control Based Reinforcement LearningCERN openlab Summer Student Lightning Talks (2/2)CERN openlab Summer Student Programme 2023<!--HTML-->Abstract: This project aims to develop a model-based Reinforcement Learning (RL) algorithm, based on the GP-MPC(Gaussian Process-Model Predictive Control) approach, for controlling accelerator systems using Machine Learning. By leveraging insufficient data and uncertainty quantification capabilities of Gaussian Processes, the algorithm will overcome limitations related to beam instrumentation and training large iterations. The final outcome will be the integration of the modified algorithm into CERN's Generic Optimisation Framework, facilitating easy deployment for operational scenarios and enabling intelligent accelerator control.oai:cds.cern.ch:28683782023
spellingShingle CERN openlab Summer Student Programme 2023
Aye, Su
Accelerator Control using Gaussian Process Model Predictive Control Based Reinforcement Learning
title Accelerator Control using Gaussian Process Model Predictive Control Based Reinforcement Learning
title_full Accelerator Control using Gaussian Process Model Predictive Control Based Reinforcement Learning
title_fullStr Accelerator Control using Gaussian Process Model Predictive Control Based Reinforcement Learning
title_full_unstemmed Accelerator Control using Gaussian Process Model Predictive Control Based Reinforcement Learning
title_short Accelerator Control using Gaussian Process Model Predictive Control Based Reinforcement Learning
title_sort accelerator control using gaussian process model predictive control based reinforcement learning
topic CERN openlab Summer Student Programme 2023
url http://cds.cern.ch/record/2868378
work_keys_str_mv AT ayesu acceleratorcontrolusinggaussianprocessmodelpredictivecontrolbasedreinforcementlearning
AT ayesu cernopenlabsummerstudentlightningtalks22

Ejemplares similares