Demonstration of system-bath physics on a gate-based quantum computer

TL;DR

This paper demonstrates a novel method leveraging qubit noise to perform nonunitary operations for algorithmic cooling, enabling the simulation of many-body quantum systems on current quantum hardware.

Contribution

It introduces a noise-based approach to implement nonunitary operations, facilitating quantum cooling and state stabilization on near-term devices.

Findings

Successfully cooled system spins to ferromagnetic and antiferromagnetic states.

Demonstrated state stabilization for systems up to three spins with four auxiliary spins.

Validated the approach on IBM-Q hardware.

Abstract

Algorithmic cooling can be used to find correlated states of many-body quantum systems. It is based on quantum circuits that perform nonunitary operations, whose implementation can be challenging on near-term quantum computers. In this work we develop a method that uses inherent qubit noise to implement nonunitary operations and algorithmic cooling. In our approach, qubit decay during quantum computation is used to simulate dissipation of auxiliary-spin bath, which cools down a simulated system towards its ground state. We test the algorithm on IBM-Q devices and demonstrate the relaxation of system spins to ferromagnetic and antiferromagnetic ordering, controlled by the definition of the system Hamiltonian. The ordering is stable as long as the algorithm is run. We are able to perform cooling and state stabilization for global systems of up to three system spins and four auxiliary spins. Our work paves the way for useful quantum simulations of many-body quantum systems on near-term quantum computers.