Normalizing Flow Methods for QCD Global Analysis in the Extraction of the Transversity PDF

QCD global analyses are at the forefront of interpreting the wealth of experimental data from observables like semi-inclusive DIS, single-inclusive proton-proton collisions, and electron-positron annihilation. In this endeavor, there is substantial computational complexity involved in analyzing the data within the theory of Quantum Chromodynamics (QCD), with existing techniques being limited in terms of efficiency and sampling accuracy. In this talk, I will present the use of Generative AI as a technique to overcome some of these computational challenges by leveraging surrogate neural networks (NNs) to address differentiable programming requirements of machine learning (ML). In particular, I will report on some preliminary work in extracting the transversity PDF from transverse single-spin asymmetries in dihadron fragmentation.