Presentations -

量子機械学習による音声異常検知の試み -工場の導入を目指した基礎 検討-

Takao Tomono, Kazuya Tsujimura

2025年第72回応用物理学会春季学術講演会, 

2025.03

, 

Oral presentation (general)

Inspection trail with factory data on quantm kernel learning

TOMONO Takao, NATSUBORI Satoko

Proceedings of the Annual Conference of JSAI, 

2023

, 

The Japanese Society for Artificial Intelligence

　View Summary

Machine learning classifiers have been used in medicine, factory inspections, and automated driving. Support Vector Machines (SVMs), one of the classifiers, are useful and have been used in various situations. In particular, kernel methods are very important for nonlinear and unsolvable classification. On the other hand, quantum machine learning has received much attention in recent years, but its specific evaluation has not been done much. In this study, we applied quantum kernel learning to the factory inspection process. As a result, it showed higher performance than classical kernel learning. This time, the image data was preprocessed, binarized, and then subjected to principal component analysis. Although the cumulative contribution rate was 75%, the accuracy was over 97% when performing quantum kernel learning. The accuracy of 93% is also obtained by classical kernel learning. Kernel learning is known to depend on the properties of the dataset, but in the future, we would like to accumulate data on what kind of datasets show the superiority of quantum.

Efficient Optimization Method Using Tensor Networks for Circuit to Generate non-Gaussian States

永井隆太郎, 友野孝夫

Optics & Photonics Japan講演予稿集(CD-ROM), 

2022

Performance evaluation of quantum kernel machine learning

TOMONO Takao, NATSUBORI Satoko, IMAIZUMI Katsumi

Proceedings of the Annual Conference of JSAI, 

2022

, 

The Japanese Society for Artificial Intelligence

　View Summary

Machine learning classifiers have been used in medicine, factory inspections, and automated driving. Support Vector Machines (SVMs), one of the classifiers, are particularly useful and have been used in various situations. In particular, kernel methods are very important for nonlinear and unsolvable classification. On the other hand, quantum machine learning has received much attention in recent years, but its specific evaluation has not been done much. In this study, we examined the process of building a learning model for classification using a heart disease data set. As a result, we found that the classical kernel method is a method to build a learning model by improving the true positive rate from a random model, while the quantum kernel method is a method to reduce the false positive rate from high true positive rate and false positive rate. In summary, we have demonstrated for the first time the process of quantum circuit learning by using the ROC space. Furthermore, we were able to construct a learning model using a quantum kernel with higher accuracy than that constructed by the classical kernel method.

Photonics quantum computing using tensor network 2: Application for real problems

永井隆太郎, 友野孝夫

応用物理学会春季学術講演会講演予稿集(CD-ROM), 

2022

Estimation of non-diagonal terms derived by VQE method using Frame Superposition Cluster (FSC)

Wakaura Hikaru, Tomono Takao

JSAP Annual Meetings Extended Abstracts, 

2021.02

, 

The Japan Society of Applied Physics

Ramer-Douglas-Peuckerアルゴリズムの量子コンピュータへの応用

湊雄一郎, 友野孝夫

応用物理学会秋季学術講演会講演予稿集(CD-ROM), 

2021

Photonics quantum computing using tensor network

永井隆太郎, 友野孝夫, 湊雄一郎

応用物理学会秋季学術講演会講演予稿集(CD-ROM), 

2021

Illumination and Reflectance Estimation with Spectral Image in Log Space

小林由枝, 川上玲, 森本哲郎, 友野孝夫, 池内克史

Optics & Photonics Japan講演予稿集(CD-ROM), 

2017

RGB画像を用いた薄膜物体の形状と色彩の推定

小林由枝, 森本哲郎, 佐藤いまり, 向川康博, 友野孝夫, 池内克史

Optics & Photonics Japan講演予稿集(CD-ROM), 

2016

ホログラム装着型ディスプレイの提案とその解像特性

友野 孝夫, 大黒 友二

三次元画像コンファレンス講演論文集, 

1999.06

, 

3次元画像コンファレンス実行委員会

PVED法により作成されたカラーフィルターの光透過特性

清水敬司, 大津茂実, 友野孝夫, あくつ英一

応用物理学会学術講演会講演予稿集, 

1999

端面照射型ホログラムの装着型ディスプレイへの適用可能性

友野孝夫, 小笠原康裕, 上柳喜一

光学シンポジウム講演予稿集, 

1999

Novel organic nonlinear optical materials containing cyclobutenedione unit.

西片康成, 友野孝夫, PU L S

応用物理学関係連合講演会講演予稿集, 

1995

Phase-matching properties of P1 crystals. Case of DAD crystals.

友野孝夫, 梅垣真祐, 宇佐川智功, PU L S, 佐々木敬介

応用物理学関係連合講演会講演予稿集, 

1994

Novel Organic Non-linear Optical Materials Containing Electron Withdrawing Cyclobutenedione-yl Group.

西片康成, 友野孝夫, PU L S

Fuji Xerox Technical Report, 

1993

Molecular structures Hyperpolarizabilities Crystal structures. Case of cyclobutenedione derivatives.

友野孝夫, 西片康成, 梅垣真祐, 佐々木敬介, PU L S

応用物理学会学術講演会講演予稿集, 

1993

Preparations and properties of phenylcycobutenedione-containing poled polymer films.

西片康成, 友野孝夫

高分子学会予稿集, 

1993

Nonlinear Optical Properties of a Novel Organic Crystal (DAD) containing Cyclobutendione as an Electron Acceptor.

友野孝夫, PU L S, 木下岳司, 佐々木敬介

応用物理学会学術講演会講演予稿集, 

1992

Crystal Structure and Second Harmonic Generation of Nonlinear Optical Materials Containing Cyclobutenedione as an Electron Acceptor.

PU L S, 友野孝夫, 堀俊彦, 城始勇

高分子学会予稿集, 

1992