Infinite-dimensional Extension of the Linear Combination of Hamiltonian Simulation: Theorems and Applications

TL;DR

This paper extends the Linear Combination of Hamiltonian Simulation to infinite-dimensional spaces, enabling simulation of complex quantum and PDE dynamics with new sampling methods and broad applicability.

Contribution

It introduces the Inf-LCHS framework for infinite-dimensional quantum simulations, including novel sampling techniques and applications to diverse non-Hermitian dynamics.

Findings

Applicable to non-Hermitian dynamics and PDEs

Provides cost estimates for quantum algorithms

Demonstrates broad applicability to quantum and classical systems

Abstract

We generalize the Linear Combination of Hamiltonian Simulation (LCHS) formula [An, Liu, Lin, Phys. Rev. Lett. 2023] to simulate time-evolution operators in infinite-dimensional spaces, including scenarios involving unbounded operators. This extension, named Inf-LCHS for short, bridges the gap between finite-dimensional quantum simulations and the broader class of infinite-dimensional quantum dynamics governed by partial differential equations (PDEs). Furthermore, we propose two sampling methods by integrating the infinite-dimensional LCHS with Gaussian quadrature schemes (Inf-LCHS-Gaussian) or Monte Carlo integration schemes (Inf-LCHS-MC). We demonstrate the applicability of the Inf-LCHS theorem to a wide range of non-Hermitian dynamics, including linear parabolic PDEs, queueing models (birth-or-death processes), Schr\"odinger equations with complex potentials, Lindblad equations, and black hole thermal field equations. Our analysis provides insights into simulating general linear dynamics using a finite number of quantum dynamics and includes cost estimates for the corresponding quantum algorithms.