Turning qubit noise into an advantage: Automatic state preparation and long-time dynamics for impurity models on quantum computers

TL;DR

This paper demonstrates that quantum noise, specifically amplitude damping, can be exploited to simulate impurity models in strongly-correlated systems more efficiently on quantum computers, reducing qubit requirements and enabling longer simulations.

Contribution

It introduces a circuit leveraging noise to simulate impurity models, reducing qubit count and eliminating the need for ground state preparation in quantum simulations.

Findings

Achieves longer time dynamics with fewer qubits.

Utilizes noise to simulate dissipative impurity models.

Eliminates the need for ground state search and preparation.

Abstract

Noise is often regarded as a limitation of quantum computers. In this work, we show that in the dynamical mean field theory (DMFT) approach to strongly-correlated systems, it can actually be harnessed to our advantage. Indeed, DMFT maps a lattice model onto an impurity model, namely a finite system coupled to a dissipative bath. While standard approaches require a large number of high-quality qubits in a unitary context, we propose a circuit that harvests amplitude damping to reproduce the dynamics of this model with a blend of noisy and noiseless qubits. We find compelling advantages with this approach: a substantial reduction in the number of qubits, the ability to reach longer time dynamics, and no need for ground state search and preparation. This method would naturally fit in a partial quantum error correction framework.