Analysis and synthesis of feature map for kernel-based quantum classifier

Yudai Suzuki; Hiroshi Yano; Qi Gao; Shumpei Uno; Tomoki Tanaka; Manato Akiyama; Naoki Yamamoto

doi:10.1007/s42484-020-00020-y

Analysis and synthesis of feature map for kernel-based quantum classifier

Yudai Suzuki, Hiroshi Yano, Qi Gao, Shumpei Uno, Tomoki Tanaka, Manato Akiyama, Naoki Yamamoto

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

17 Citations (Scopus)

Abstract

A method for analyzing the feature map for the kernel-based quantum classifier is developed; that is, we give a general formula for computing a lower bound of the exact training accuracy, which helps us to see whether the selected feature map is suitable for linearly separating the dataset. We show a proof of concept demonstration of this method for a class of 2-qubit classifier, with several 2-dimensional datasets. Also, a synthesis method, which combines different kernels to construct a better-performing feature map in a lager feature space, is presented.

Original language	English
Journal	Quantum Machine Intelligence
Volume	2
Issue number	1
DOIs	https://doi.org/10.1007/s42484-020-00020-y
Publication status	Published - 2020 Jun 1

Keywords

Feature map
Kernel method
Quantum computing
Support vector machine

ASJC Scopus subject areas

Artificial Intelligence
Computational Theory and Mathematics
Software
Applied Mathematics
Theoretical Computer Science

Access to Document

10.1007/s42484-020-00020-y

Cite this

@article{a75572066dc34ce99626df3623e56610,

title = "Analysis and synthesis of feature map for kernel-based quantum classifier",

abstract = "A method for analyzing the feature map for the kernel-based quantum classifier is developed; that is, we give a general formula for computing a lower bound of the exact training accuracy, which helps us to see whether the selected feature map is suitable for linearly separating the dataset. We show a proof of concept demonstration of this method for a class of 2-qubit classifier, with several 2-dimensional datasets. Also, a synthesis method, which combines different kernels to construct a better-performing feature map in a lager feature space, is presented.",

keywords = "Feature map, Kernel method, Quantum computing, Support vector machine",

author = "Yudai Suzuki and Hiroshi Yano and Qi Gao and Shumpei Uno and Tomoki Tanaka and Manato Akiyama and Naoki Yamamoto",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Nature Switzerland AG.",

year = "2020",

month = jun,

day = "1",

doi = "10.1007/s42484-020-00020-y",

language = "English",

volume = "2",

journal = "Quantum Machine Intelligence",

issn = "2524-4906",

publisher = "Springer Science and Business Media Deutschland GmbH",

number = "1",

}

TY - JOUR

T1 - Analysis and synthesis of feature map for kernel-based quantum classifier

AU - Suzuki, Yudai

AU - Yano, Hiroshi

AU - Gao, Qi

AU - Uno, Shumpei

AU - Tanaka, Tomoki

AU - Akiyama, Manato

AU - Yamamoto, Naoki

PY - 2020/6/1

Y1 - 2020/6/1

N2 - A method for analyzing the feature map for the kernel-based quantum classifier is developed; that is, we give a general formula for computing a lower bound of the exact training accuracy, which helps us to see whether the selected feature map is suitable for linearly separating the dataset. We show a proof of concept demonstration of this method for a class of 2-qubit classifier, with several 2-dimensional datasets. Also, a synthesis method, which combines different kernels to construct a better-performing feature map in a lager feature space, is presented.

AB - A method for analyzing the feature map for the kernel-based quantum classifier is developed; that is, we give a general formula for computing a lower bound of the exact training accuracy, which helps us to see whether the selected feature map is suitable for linearly separating the dataset. We show a proof of concept demonstration of this method for a class of 2-qubit classifier, with several 2-dimensional datasets. Also, a synthesis method, which combines different kernels to construct a better-performing feature map in a lager feature space, is presented.

KW - Feature map

KW - Kernel method

KW - Quantum computing

KW - Support vector machine

UR - http://www.scopus.com/inward/record.url?scp=85092767576&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85092767576&partnerID=8YFLogxK

U2 - 10.1007/s42484-020-00020-y

DO - 10.1007/s42484-020-00020-y

M3 - Article

AN - SCOPUS:85092767576

SN - 2524-4906

VL - 2

JO - Quantum Machine Intelligence

JF - Quantum Machine Intelligence

IS - 1

ER -

Analysis and synthesis of feature map for kernel-based quantum classifier

Abstract

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this