Designing complex concentrated alloys with quantum machine learning and language modeling

Zongrui Pei; Yilun Gong; Xianglin Liu; Junqi Yin

doi:10.1016/j.matt.2024.05.035

Designing complex concentrated alloys with quantum machine learning and language modeling

Zongrui Pei
, Yilun Gong
, Xianglin Liu
, Junqi Yin

Analytics and AI Methods at Scale

Research output: Contribution to journal › Article › peer-review

8 Scopus citations

Abstract

Designing novel complex concentrated alloys (CCAs) is an essential topic in materials science. However, due to the complicated high-dimensional component-property relationship, tuning material properties by researchers’ experience is challenging, even when guided by physical or empirical rules. Here, we adopt quantum computing (QC) technology and machine learning models to provide a proof-of-concept application of QC in physical metallurgy. We propose a quantum support vector machine (QSVM) model to predict single-phase CCAs. We show that fine-tuned quantum kernels with entanglement deliver promising performance, with a maximum accuracy of 89.4%. The QSVM model is then used to identify 1,741 lightweight CCAs jointly with a new text-mining-based method. Meanwhile, we devise a controllable approach to study the effect of noise on model performance and find that the noise level needs to be minimized for high-performance QSVM models. This study provides a practical and general approach to designing CCAs based on quantum technologies.

Original language	English
Pages (from-to)	3433-3446
Number of pages	14
Journal	Matter
Volume	7
Issue number	10
DOIs	https://doi.org/10.1016/j.matt.2024.05.035
State	Published - Oct 2 2024

Funding

This work was supported in part through the NYU IT High Performance Computing resources, services, and staff expertise. We also acknowledge the use of IBM Quantum services for this work. The views expressed are those of the authors and do not reflect the official policy or position of IBM or the IBM Quantum team. Contributions to this work by X.L. were completed before the author joined Amazon. Z.P. conceived the project and designed the study. Z.P. collected data, wrote the code, and trained the initial version of ML models. Y.G. performed and finalized the ML models used in this article and wrote code for post-processing data. Y.G. and J.Y. performed tests that placed the results on a solid basis. X.L. analyzed the data and helped prepare the figures. All authors contributed to the discussion and finalized the manuscript. The authors declare no competing interests.

Keywords

Alloy design
MAP 2: Benchmark
language models
quantum computer
quantum machine learning

Access to Document

10.1016/j.matt.2024.05.035

Cite this

@article{4bca2d3272894871ae3a32430a7973ba,

title = "Designing complex concentrated alloys with quantum machine learning and language modeling",

abstract = "Designing novel complex concentrated alloys (CCAs) is an essential topic in materials science. However, due to the complicated high-dimensional component-property relationship, tuning material properties by researchers{\textquoteright} experience is challenging, even when guided by physical or empirical rules. Here, we adopt quantum computing (QC) technology and machine learning models to provide a proof-of-concept application of QC in physical metallurgy. We propose a quantum support vector machine (QSVM) model to predict single-phase CCAs. We show that fine-tuned quantum kernels with entanglement deliver promising performance, with a maximum accuracy of 89.4\%. The QSVM model is then used to identify 1,741 lightweight CCAs jointly with a new text-mining-based method. Meanwhile, we devise a controllable approach to study the effect of noise on model performance and find that the noise level needs to be minimized for high-performance QSVM models. This study provides a practical and general approach to designing CCAs based on quantum technologies.",

keywords = "Alloy design, MAP 2: Benchmark, language models, quantum computer, quantum machine learning",

author = "Zongrui Pei and Yilun Gong and Xianglin Liu and Junqi Yin",

note = "Publisher Copyright: {\textcopyright} 2024 Elsevier Inc.",

year = "2024",

month = oct,

day = "2",

doi = "10.1016/j.matt.2024.05.035",

language = "English",

volume = "7",

pages = "3433--3446",

journal = "Matter",

issn = "2590-2393",

publisher = "Cell Press",

number = "10",

}

TY - JOUR

T1 - Designing complex concentrated alloys with quantum machine learning and language modeling

AU - Pei, Zongrui

AU - Gong, Yilun

AU - Liu, Xianglin

AU - Yin, Junqi

PY - 2024/10/2

Y1 - 2024/10/2

N2 - Designing novel complex concentrated alloys (CCAs) is an essential topic in materials science. However, due to the complicated high-dimensional component-property relationship, tuning material properties by researchers’ experience is challenging, even when guided by physical or empirical rules. Here, we adopt quantum computing (QC) technology and machine learning models to provide a proof-of-concept application of QC in physical metallurgy. We propose a quantum support vector machine (QSVM) model to predict single-phase CCAs. We show that fine-tuned quantum kernels with entanglement deliver promising performance, with a maximum accuracy of 89.4%. The QSVM model is then used to identify 1,741 lightweight CCAs jointly with a new text-mining-based method. Meanwhile, we devise a controllable approach to study the effect of noise on model performance and find that the noise level needs to be minimized for high-performance QSVM models. This study provides a practical and general approach to designing CCAs based on quantum technologies.

AB - Designing novel complex concentrated alloys (CCAs) is an essential topic in materials science. However, due to the complicated high-dimensional component-property relationship, tuning material properties by researchers’ experience is challenging, even when guided by physical or empirical rules. Here, we adopt quantum computing (QC) technology and machine learning models to provide a proof-of-concept application of QC in physical metallurgy. We propose a quantum support vector machine (QSVM) model to predict single-phase CCAs. We show that fine-tuned quantum kernels with entanglement deliver promising performance, with a maximum accuracy of 89.4%. The QSVM model is then used to identify 1,741 lightweight CCAs jointly with a new text-mining-based method. Meanwhile, we devise a controllable approach to study the effect of noise on model performance and find that the noise level needs to be minimized for high-performance QSVM models. This study provides a practical and general approach to designing CCAs based on quantum technologies.

KW - Alloy design

KW - MAP 2: Benchmark

KW - language models

KW - quantum computer

KW - quantum machine learning

UR - https://www.scopus.com/pages/publications/85198219892

U2 - 10.1016/j.matt.2024.05.035

DO - 10.1016/j.matt.2024.05.035

M3 - Article

AN - SCOPUS:85198219892

SN - 2590-2393

VL - 7

SP - 3433

EP - 3446

JO - Matter

JF - Matter

IS - 10

ER -

Designing complex concentrated alloys with quantum machine learning and language modeling

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this