Accelerating Quantum Eigensolver Algorithms With Machine Learning

Avner Bensoussan; Elena Chachkarova; Karine Even-Mendoza; Sophie Fortz; Connor Lenihan

Accelerating Quantum Eigensolver Algorithms With Machine Learning

Avner Bensoussan, Elena Chachkarova, Karine Even-Mendoza, Sophie Fortz, Connor Lenihan

Research output: Working paper/Preprint › Preprint

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.

Original language	English
Place of Publication	online
Publisher	arXiv
Number of pages	22
Volume	CoRR abs/2409.13587
Publication status	Published - 20 Sept 2024

Access to Document

https://arxiv.org/abs/2409.13587Licence: Other

Cite this

@techreport{ba68b7650a144a099baab65b62d5a555,

title = "Accelerating Quantum Eigensolver Algorithms With Machine Learning",

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{\textquoteright}s Pythonregressor. We evaluated our preliminary approach on 20-, 24- and 28-qubit systems by optimising the Eigensolver{\textquoteright}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 onHamiltonian characteristics. In future work, we plan to train machine learning models to optimise other aspects or subroutines of quantum algorithm execution beyond its hyperparameters.",

author = "Avner Bensoussan and Elena Chachkarova and Karine Even-Mendoza and Sophie Fortz and Connor Lenihan",

year = "2024",

month = sep,

day = "20",

language = "English",

volume = "CoRR abs/2409.13587",

publisher = "arXiv",

type = "WorkingPaper",

institution = "arXiv",

}

TY - UNPB

T1 - Accelerating Quantum Eigensolver Algorithms With Machine Learning

AU - Bensoussan, Avner

AU - Chachkarova, Elena

AU - Even-Mendoza, Karine

AU - Fortz, Sophie

AU - Lenihan, Connor

PY - 2024/9/20

Y1 - 2024/9/20

N2 - 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 Pythonregressor. 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 onHamiltonian characteristics. In future work, we plan to train machine learning models to optimise other aspects or subroutines of quantum algorithm execution beyond its hyperparameters.

AB - 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 Pythonregressor. 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 onHamiltonian characteristics. In future work, we plan to train machine learning models to optimise other aspects or subroutines of quantum algorithm execution beyond its hyperparameters.

M3 - Preprint

VL - CoRR abs/2409.13587

BT - Accelerating Quantum Eigensolver Algorithms With Machine Learning

PB - arXiv

CY - online

ER -

Accelerating Quantum Eigensolver Algorithms With Machine Learning

Abstract

Access to Document

Other files and links

Fingerprint

AccelerQ: Accelerating Quantum Eigensolvers With Machine Learning on Quantum Simulators

Artifact of Accelerating Quantum Algorithms With Machine Learning

Cite this

Accelerating Quantum Eigensolver Algorithms With Machine Learning

Abstract

Access to Document

Other files and links

Fingerprint

Research output

AccelerQ: Accelerating Quantum Eigensolvers With Machine Learning on Quantum Simulators

Datasets

Artifact of Accelerating Quantum Algorithms With Machine Learning

Cite this