Machine Learning Approaches in Lattice QCD

Lattice QCD underpins precision calculations of strong-interaction effects needed for Standard Model tests, flavour physics and searches for physics beyond the Standard Model. Progress in these areas increasingly depends on overcoming computational bottlenecks that cannot be addressed through hardware improvements alone. As a means of addressing these challenges, machine learning methods are gaining increasing traction in the Lattice QCD workflow. As these methods move beyond proof-of-concept demonstrations towards at-scale applications, it is necessary to review progress, evaluate promising applications, and develop a roadmap for continued impacts on the field of Lattice QCD.

The third edition of Machine Learning Approaches in Lattice QCD will be hosted by the Higgs Centre for Theoretical Physics at the University of Edinburgh. This five-day meeting will bring together researchers at the forefront of this field through a plenary talk schedule, discussion sessions, a poster session, and plenty of opportunity for informal interactions. Building on the first two editions in this series, this workshop will examine recent methodological progress and lessons from practical applications. Main topics to be covered in this workshop include:

• Flow and diffusion models
• Inverse problems
• Observable estimation
• Neural network quantum states
• Physics-informed architectures
• Scaling and practical implementation

The workshop will take place on May 10-14, 2027 — please save the date. Registration and further details will be available soon.

All participants in HCTP activities are expected to show respect and courtesy to others throughout the duration of an event, including at social events. We draw the participants’ attention to the HCTP Code of Conduct.