arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2609.38711nucl-thnucl-ex

利用机器学习和人工智能推进核物理研究

Advancing Nuclear Physics with Machine Learning and Artificial Intelligence

Wanbing He, Qingfeng Li, Yugang Ma, Zhongming Niu, Junchen Pei, Yingxun Zhang

首次发表
浏览论文内容

中文总结 AI 辅助

本文综述了机器学习与人工智能在核物理中的最新进展,涵盖核结构、重离子碰撞及多体波函数等方向,强调从数据推断向物理信息学习演变,并展望未来融合方向。

中文摘要 AI 辅助

机器学习(ML)和人工智能(AI)正成为各学科科学研究中的强大工具。在这篇综述中,我们总结了AI辅助研究的最新进展,涵盖核结构与反应观测量、重离子碰撞与致密核物质性质、多体波函数、实验设施及数据分析等方面。本综述重点关注自2023年上一篇综述以来的新进展,核物理中的机器学习正从纯粹的数据推断演变为物理信息驱动的学习。我们还综述了将物理知识与现代学习架构、大型基础模型及其他新兴方法相结合的未来方向。这些进展表明,AI正在赋能并推进应对最具挑战性核物理问题的新方法。

英文摘要

Machine learning (ML) and artificial intelligence (AI) are becoming powerful tools in scientific research across various disciplines. In this review, we summarize recent progress in AI-assisted studies of nuclear structure and reaction observables, heavy-ion collisions and dense nuclear matter properties, many-body wave functions, experimental facilities and data analysis. This review focus on new progress since the last review in 2023, and machine learning in nuclear physics is evolving from purely data inferences to physics informed learning. Future directions on the integration of physical knowledge with modern learning architectures, large foundation models and other emerging methods are also reviewed. These developments suggest that AI is enabling and advancing new approaches towards most challenging nuclear physics problems.

发表机构

  • Fudan University(复旦大学)
  • Shanghai Research Center for Theoretical Nuclear Physics, NSFC and Fudan University(国家自然科学基金委与复旦大学理论核物理上海研究中心)
  • Huzhou Normal University(湖州师范学院)
  • China Institute of Atomic Energy(中国原子能科学研究院)
  • East China Normal University(华东师范大学)
  • Anhui University(安徽大学)
  • Peking University(北京大学)
  • Chinese Academy of Sciences(中国科学院)
  • Guangxi Normal University(广西师范大学)

机构由 AI 辅助整理,请以论文原文为准。

补充信息

↑