Prof. Simonetta Liuti (University of Virginia)

"Obtaining 3D images of the ocean quarks and gluons at the Electron Ion Collider with Artificial Intelligence"

Abstract: AI’s capacity to solve complex scientific and engineering problems has been recently growing exponentially in all its standard subfields, including natural language processing and generation, imaging, multi-agent systems, and motor control for robotics. Physics plays a unique role in this context: on one side, integrating physics laws and their known symmetries and mathematical constraints with machine learning methodologies provides a fertile ground for finding solutions to problems that would be otherwise intractable. On the other hand, physics reasoning patterns can be used to understand the working of machine learning algorithms.

This talk focuses on the area of nuclear and particle physics. I will address, in particular, the exploration of physics-guided ML frameworks to extract essential information on the QCD structure of strongly interacting systems from experimental data at both current facilities and at the new Electron Ion Collider (EIC).

Bio: Professor Liuti´s research involves theoretical studies of the quark and gluon internal dynamics of the proton. The fundamental interactions of quarks and gluons are governed by the beautifully concise Lagrangian of quantum chromodynamics (QCD), however, to be able to understand the mechanisms generating the proton mass, spin and mechanical properties in terms of quarks and gluons is one of the biggest challenges in physics today. Professor Liuti's work has been focused on understanding how the nucleon gets its spin from its fundamental degrees of freedom: is the orbital motion of quarks and gluons relevant and how do we describe it in QCD? The theoretical tools that allow us to obtain information on angular momentum and other mechanical properties including pressure and shear are the 3D parton distributions, or the Wigner distributions at the femtoscale. A new generation of current and planned experiments at the future Electron Ion Collider (EIC) could in principle allow us to incorporate all the information from data and phenomenology into a tomographic image connecting the deepest part of the quantum world with what we see as everyday matter around us. Accessing this new type of knowledge has far reaching consequences as it could even impact our understanding of the interior of neutron stars.