Accelerating Quantum Eigensolver Algorithms With Machine Learning

TL;DR

This paper investigates using machine learning to accelerate quantum eigensolver algorithms on NISQ devices, demonstrating potential error reduction in large systems through hyperparameter prediction.

Contribution

It introduces a machine learning approach to optimize hyperparameters in quantum eigensolvers, showing initial success on large qubit systems and proposing further research directions.

Findings

0.12% error reduction in 28-qubit systems

Preliminary models trained on systems up to 16 qubits

Inconclusive results for 20- and 24-qubit systems

Abstract

In this paper, we explore accelerating Hamiltonian ground state energy calculation on NISQ devices. We suggest using search-based methods together with machine learning to accelerate quantum algorithms, exemplified in the Quantum Eigensolver use case. We trained two small models on classically mined data from systems with up to 16 qubits, using XGBoost's Python regressor. We evaluated our preliminary approach on 20-, 24- and 28-qubit systems by optimising the Eigensolver's hyperparameters. These models predict hyperparameter values, leading to a 0.12% reduction in error when tested on 28-qubit systems. However, due to inconclusive results with 20- and 24-qubit systems, we suggest further examination of the training data based on Hamiltonian characteristics. In future work, we plan to train machine learning models to optimise other aspects or subroutines of quantum algorithm execution beyond its hyperparameters.