Artificial Intelligence (AI) and Machine learning (ML) promise significant enhancements for particle accelerator operations, including applications in diagnostics, controls, and modeling. Challenges still exist in experimentally verifying AI/ML methods before deployment at user facilities. The ability to quickly generalize and adapt these methods to new operating configurations at the same facility or between facilities also remains a challenge and requires combining model-independent adaptive feedback with traditional ML tools.
These methods also apply to the detection, classification, and prevention of operational anomalies that can cause accelerator damage or excessive beam loss in the case of abnormal operations. Opportunity exists in broadening AI/ML methods for early detection of a broad range of accelerator component or subsystem failures.
Modern accelerator design requires the optimization of large numbers of coupled accelerator parameters, through advanced numerical simulations. AI/ML methods can be used to speed up identification of the most promising combinations of parameters and thereby reduce the total number of required simulations.
The goal of this new research topic is to promote awareness and broader application of Artificial Intelligence and Machine Learning to problems in accelerator physics, simulation and design, and optimization. Application of these techniques has increased in recent years. Additionally, their value in understanding accelerator beam properties and accelerator performance beyond the standard methods has become clear, as well as becoming key to making significant advances in machine optimization and performance, often when minimal measurement data are available.
We are seeking contributions covering all aspects of accelerator theory, design, and operations. Contributions may include reports of original research, application of methods including experimental results, and reviews of general or focused aspects of Artificial Intelligence (AI) and Machine Learning (ML) as applied to accelerators.
Topics of particular interest may include (but not limited to):
• Accelerator ML/AI basics
• New or novel ideas on ML/AI theory and computational algorithms
• Application to accelerator diagnostics
• Accelerator operations optimization
• Application to accelerator system fault diagnosis and correction
• Application to accelerator system design
Dr. Garnett is a part-time consultant for TechSource, Inc. All other Topic Editors declare no competing interests with regards to the Research Topic subject.
Artificial Intelligence (AI) and Machine learning (ML) promise significant enhancements for particle accelerator operations, including applications in diagnostics, controls, and modeling. Challenges still exist in experimentally verifying AI/ML methods before deployment at user facilities. The ability to quickly generalize and adapt these methods to new operating configurations at the same facility or between facilities also remains a challenge and requires combining model-independent adaptive feedback with traditional ML tools.
These methods also apply to the detection, classification, and prevention of operational anomalies that can cause accelerator damage or excessive beam loss in the case of abnormal operations. Opportunity exists in broadening AI/ML methods for early detection of a broad range of accelerator component or subsystem failures.
Modern accelerator design requires the optimization of large numbers of coupled accelerator parameters, through advanced numerical simulations. AI/ML methods can be used to speed up identification of the most promising combinations of parameters and thereby reduce the total number of required simulations.
The goal of this new research topic is to promote awareness and broader application of Artificial Intelligence and Machine Learning to problems in accelerator physics, simulation and design, and optimization. Application of these techniques has increased in recent years. Additionally, their value in understanding accelerator beam properties and accelerator performance beyond the standard methods has become clear, as well as becoming key to making significant advances in machine optimization and performance, often when minimal measurement data are available.
We are seeking contributions covering all aspects of accelerator theory, design, and operations. Contributions may include reports of original research, application of methods including experimental results, and reviews of general or focused aspects of Artificial Intelligence (AI) and Machine Learning (ML) as applied to accelerators.
Topics of particular interest may include (but not limited to):
• Accelerator ML/AI basics
• New or novel ideas on ML/AI theory and computational algorithms
• Application to accelerator diagnostics
• Accelerator operations optimization
• Application to accelerator system fault diagnosis and correction
• Application to accelerator system design
Dr. Garnett is a part-time consultant for TechSource, Inc. All other Topic Editors declare no competing interests with regards to the Research Topic subject.