Novel Algorithms for Computing Correlation Functions of Nuclei

TL;DR

This paper introduces two new algorithms that leverage permutation symmetry in tensor expressions to significantly reduce the computational cost of calculating nuclear correlation functions, enabling more feasible ab initio studies of larger nuclei.

Contribution

The paper presents novel algorithms that exploit permutation symmetry in tensor expressions to speed up nuclear correlation function computations.

Findings

Significant speed-ups achieved for certain interpolating operators.

Reduction in computational complexity for nuclear correlation functions.

Potential to enable ab initio studies of larger nuclei.

Abstract

The computational cost required to calculate nuclear correlation functions grows factorially in the number of quarks, making the study of large nuclei inaccessible to ab initio study using lattice QCD at the present time. However, the tensor expressions corresponding to many of these correlation functions exhibit a high degree of permutation symmetry that can be exploited to reduce computational work. We present promising speed-ups for certain choices of interpolating operators using two new algorithms for computing nuclear correlation functions.