Dynamic thermalization on noisy quantum hardware

TL;DR

This paper demonstrates that thermal observables and fluctuations can be obtained on small closed quantum systems without a thermal bath by averaging randomized quantum evolutions, enabling thermal physics studies on noisy quantum hardware.

Contribution

It introduces a method to simulate thermal properties on small quantum systems without a thermal bath by averaging over randomized evolutions, validated on IBM quantum hardware.

Findings

Thermal occupation probabilities with finite positive and negative temperatures were experimentally observed.

Averaging over random evolutions helps mitigate errors and incorporates noise effects into the simulated temperature.

The approach enables probing equilibrium properties at finite temperatures on noisy intermediate-scale quantum devices.

Abstract

Emulating thermal observables on a digital quantum computer is essential for quantum simulation of many-body physics. However, thermalization typically requires a large system size due to incorporating a thermal bath, whilst limited resources of near-term digital quantum processors allow for simulating relatively small systems. We show that thermal observables and fluctuations may be obtained for a small closed system without a thermal bath. Thermal observables occur upon classically averaging quantum mechanical observables over randomized variants of their time evolution that run independently on a digital quantum processor. Using an IBM quantum computer, we experimentally find thermal occupation probabilities with finite positive and negative temperatures defined by the initial state's energy. Averaging over random evolutions facilitates error mitigation, with the noise contributing to the temperature in the simulated observables. This result fosters probing the dynamical emergence of equilibrium properties of matter at finite temperatures on noisy intermediate-scale quantum hardware.