A Tour of Tools, Trends, and Tradeoffs in AI/ML for High Energy Physics

Speakers: Dr Shawn Dubey (Brown University)

Machine Learning (ML) has advanced rapidly over the past decade, transforming the way data-intensive sciences are conducted. High energy physics is distinctive among the physical sciences in its early adoption and deployment of new AI/ML methods, driven by extreme data volumes, complex event structures, and stringent real-time constraints.

In this talk, I will trace the evolution of these approaches through my own research trajectory, beginning with the use of multilayer perceptrons and physics-motivated feature engineering, and progressing toward modern techniques that emphasize data curation, weak supervision, and hardware–software co-design of AI/ML algorithms. I will highlight cutting edge methods and how they arise from the interaction between models, data, and experimental constraints, and how this perspective informs future research directions.

https://indico.phys.hawaii.edu/event/2769/