S-matrix Bootstrap with machine learning

by Damien Leflot (LAPTh)

Monday 27 Apr 2026, 14:30 → 16:30 Europe/Rome

The S-matrix bootstrap uses general physical principles like unitarity, analyticity, and crossing symmetry to study strongly interacting dynamics beyond perturbation theory. While the modern bootstrap has achieved significant success, I will motivate the need for a new framework that explicitly incorporates both elastic and inelastic unitarity. I will then describe such a framework: the Atkinson-Mandelstam approach and explain how deep-learning techniques enable its practical realization. Finally, I will present some of the results obtained using this approach.