Approximating Kaon Structure Data with Machine Learning

Quantum Chromodynamics (QCD), is the study of quark gluon interactions, and the strong nuclear force. There are currently many experiments and studies being conducted to try to further the understanding of QCD and the structure of subatomic particles. One of the particles of interest in a kaon (K). A kaon is a particle that consists of an anti-stange and up quark pair. Kaons are of particular interest because they are unstable, resulting in a very quick decay. The quick decay makes kaons very difficult to study experimentally. Fortunately, vital information regarding particle structure can be gained through studying Lattice QCD correlator functions. These functions can be quite computationally expensive to solve for, and for this reason machine learning techniques were tested to estimate additional correlator data. Machine Learning is a computational method that utilizes a data set to make a model that can predict further data. Specifically in this work, machine learning methods were examined and refined in order to predict later distance and greater momentum kaon correlator data with earlier data points.