Unveiling the pole structure of S-matrix using deep learning

TL;DR

This paper introduces a deep learning approach to analyze the pole structure of the S-matrix in particle scattering, specifically targeting inelastic processes and resonance identification in coupled channel scattering.

Contribution

It develops a curriculum learning-based neural network method to extract pole configurations from scattering data, addressing a key challenge in understanding subatomic phenomena.

Findings

Effective extraction of pole configurations in $$ scatterings

Demonstrates the method's potential for analyzing complex coupled channel problems

Provides a new tool for studying resonance phenomena in particle physics

Abstract

Particle scattering is a powerful tool to unveil the nature of various subatomic phenomena. The key quantity is the scattering amplitude whose analytic structure carries the information of the quantum states. In this work, we demonstrate our first step attempt to extract the pole configuration of inelastic scatterings using the deep learning method. Among various problems, motivated by the recent new hadron phenomena, we develop a curriculum learning method of deep neural network to analyze coupled channel scattering problems. We show how effectively the method works to extract the pole configuration associated with resonances in the $\pi N$ scatterings.