Machine learning as a discovery tool in hep-th

Abstract: Machine learning provides a new tool for analyzing Big Data and Small Data in mathematics and theoretical physics. In this talk, I discuss two case studies. The first predicts the volume of the knot complement of hyperbolic knots from the Jones polynomial. The second predicts the masses of baryons such as the proton and neutron from knowledge only of the meson spectrum and distinguishes between different composition hypotheses for exotic QCD resonances. Both investigations point to the existence of new analytic formulae.

HEP - theory

Audience: researchers in the topic

Theoretical Physics Seminars (Kentucky)