Machine Learning and Artificial Intelligence in Hadronic Physics

Machine learning (ML) and artificial intelligence (AI) techniques are now actively used in multiple aspects of hadronic physics, both theoretically and experimentally, the latter being the focus of this talk. ML and AI target practically all facets of QCD theory and contribute to enhancing the precision of measurements in hadronic physics, detecting the nature of measured events, explaining the underlying mechanisms from final distribution of particles, and speeding up theoretical calculations. In experiments, ML/AI are largely used for particle reconstruction, identification, event classification, and spectroscopy, and will be used in nearly every system of future QCD frontier experiments, such as the Electron Ion Collider; AI is also increasingly being used for autonomous control and experimentation, present experiment operations and calibrations, and the design of future QCD experiments. The development of next-generation data acquisition systems that stream all data from each detector to a data center to be analyzed, tagged, and filtered can further increase the implementation of ML/AI-based solutions for near real-time analysis, resulting in a shorter time to produce scientific results. This talk will provide you with a global yet non-exhaustive picture of ML/AI activities in experimental hadronic physics, the needs highlighted by our community to leverage ML/AI for QCD research, as well as potential future prospects for ML/AI applications in hadronic physics.