A Renormalisation-Group Algorithm for Eigenvalue Density Functions of Interacting Quantum Systems

TL;DR

This paper introduces a certifiable, highly accurate algorithm based on renormalization group techniques to compute the eigenvalue density function of 1D interacting quantum spin systems, including disordered and critical cases.

Contribution

It presents a novel algorithm that uses matrix product states to accurately approximate eigenvalue densities and analyze ground state properties in complex quantum systems.

Findings

Accurately approximates eigenvalue density functions for 1D quantum systems.

Enables estimation of ground-state energy gaps with confidence intervals.

Applicable to disordered and critical quantum spin models.

Abstract

We present a certifiable algorithm to calculate the eigenvalue density function -- the number of eigenvalues within an infinitesimal interval -- for an arbitrary 1D interacting quantum spin system. Our method provides an arbitrarily accurate numerical representation for the smeared eigenvalue density function, which is the convolution of the eigenvalue density function with a gaussian of prespecified width. In addition, with our algorithm it is possible to investigate the density of states near the ground state. This can be used to numerically determine the size of the ground-state energy gap for the system to within a prespecified confidence interval. Our method exploits a finitely correlated state/matrix product state representation of the propagator and applies equally to disordered and critical interacting 1D quantum spin systems. We illustrate our method by calculating an approximation to the eigenvalue density function for a random antiferromagnetic Heisenberg model.