Machine learning and artificial intelligence (ML/AI) have achieved significant advances and demonstrated the potential to achieve breakthroughs in scientific discoveries across many areas. Similarly, High-Performance Computing (HPC) systems have been utilized as computing engines for simulations for many years across many sciences and engineering disciplines, which is now under the spotlight of computing engines for ML/AI tasks. Thus, we invite researchers and practitioners to contribute to this theme of machine learning in high-performance computing environments for High Energy Physics (HEP), to discuss challenges and opportunities related to utilizing HPC systems for machine learning, such as exploiting data parallelism, model parallelism, parameter search, utilizing emerging new accelerators, and enhancing the deployment of computing resources at scale, while focusing on high energy physics or closely relevant areas.
The major topics of interest to this collection include but are not limited to:
• Novel AI/ML methods for high energy physics, focusing on data-intensive, large-scale, distributed modeling and learning, utilizing high-performance computing or emerging HPC computing environments.
• Enhancing the applicability of machine learning for high energy physics in HPC environments (e.g., feature detection, feature engineering, usability, explainability, robustness, and uncertainty quantification)
• Optimized training of machine learning models on large high energy physics data, either from simulations or experiments
• Machine learning enhanced modeling and simulation of high energy physics problems.
• Novel methods to utilize emerging hardware such as neuromorphic processors, to accelerate AI/ML models for high energy physics data in HPC environments.
• Overcoming the problems inherent to large datasets (e.g., noisy labels, missing data, scalable ingest) in high energy physics problems.
Keywords:
Machine Learning, High Energy Physics, High-Performance Computing
Important Note:
All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.
Machine learning and artificial intelligence (ML/AI) have achieved significant advances and demonstrated the potential to achieve breakthroughs in scientific discoveries across many areas. Similarly, High-Performance Computing (HPC) systems have been utilized as computing engines for simulations for many years across many sciences and engineering disciplines, which is now under the spotlight of computing engines for ML/AI tasks. Thus, we invite researchers and practitioners to contribute to this theme of machine learning in high-performance computing environments for High Energy Physics (HEP), to discuss challenges and opportunities related to utilizing HPC systems for machine learning, such as exploiting data parallelism, model parallelism, parameter search, utilizing emerging new accelerators, and enhancing the deployment of computing resources at scale, while focusing on high energy physics or closely relevant areas.
The major topics of interest to this collection include but are not limited to:
• Novel AI/ML methods for high energy physics, focusing on data-intensive, large-scale, distributed modeling and learning, utilizing high-performance computing or emerging HPC computing environments.
• Enhancing the applicability of machine learning for high energy physics in HPC environments (e.g., feature detection, feature engineering, usability, explainability, robustness, and uncertainty quantification)
• Optimized training of machine learning models on large high energy physics data, either from simulations or experiments
• Machine learning enhanced modeling and simulation of high energy physics problems.
• Novel methods to utilize emerging hardware such as neuromorphic processors, to accelerate AI/ML models for high energy physics data in HPC environments.
• Overcoming the problems inherent to large datasets (e.g., noisy labels, missing data, scalable ingest) in high energy physics problems.
Keywords:
Machine Learning, High Energy Physics, High-Performance Computing
Important Note:
All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.