Model generation method, data presentation method, data generation method, inference method, model generation device, data presentation device, data generation device, and inference device

Patent metadata

Field	Value
Publication number	US-2024257922-A1
Application number	US-202218686602-A
Country	US
Kind code	A1
Filing date	Aug 17, 2022
Priority date	Sep 27, 2021
Publication date	Aug 1, 2024
Grant date	—

Abstract

A model generation method according to one aspect of the present invention acquires first data and second data regarding a crystal structure of a material, and performs machine learning for a first encoder and a second encoder by using the first data and the second data. The second data indicates a property of the material with an index different from that of the first data. The first encoder is configured to convert the first data into a first feature vector, and the second encoder is configured to convert the second data into a second feature vector. The dimension of the first feature vector is the same as the dimension of the second feature vector. In machine learning, the first encoder and the second encoder are trained so that the values of the feature vectors of the positive samples are positioned close to each other, and the feature vector of the negative sample is positioned far from the feature vector of the positive sample.

First claim

1 . A model generation method comprising:
a step of acquiring, by a computer, first data and second data regarding a crystal structure of a material, in which the second data indicates a property of the material with an index different from an index of the first data, the first data and the second data acquired include a positive sample and a negative sample, the positive sample includes a combination of first data and second data for the same material, and the negative sample includes at least one of first data and second data for a material different from the material of the positive sample; and a step of performing, by the computer, machine learning for a first encoder and a second encoder by using the first data and the second data acquired, in which the first encoder is configured to convert the first data into a first feature vector, the second encoder is configured to convert the second data into a second feature vector, a dimension of the first feature vector is the same as a dimension of the second feature vector, and the machine learning for the first encoder and the second encoder is configured by training the first encoder and the second encoder so that values of a first feature vector and a second feature vector calculated from the first data and the second data of the positive sample are positioned close to each other, and a value of at least one of a first feature vector and a second feature vector calculated from at least one of the first data and the second data of the negative sample is positioned far from a value of at least one of the first feature vector and the second feature vector calculated from the positive sample.

2 . The model generation method according to claim 1 , further comprising the step of performing, by the computer, machine learning for a first decoder,
wherein the machine learning for the first decoder is configured by training the first decoder so that a result of the first decoder restoring the first data from a first feature vector, calculated from the first data by using the first encoder, matches the first data.

3 . The model generation method according to claim 1 , further comprising the step of performing, by the computer, machine learning for a second decoder,
wherein the machine learning for the second decoder is configured by training the second decoder so that a result of restoring the second data by the second decoder from a second feature vector, calculated from the second data by using the second encoder, matches the second data.

4 . The model generation method according to claim 1 , further comprising the step of performing, by the computer, machine learning for an estimator,
wherein in the step of acquiring the first data and the second data, the computer further acquires correct information indicating a characteristic of the material, and the machine learning for the estimator is configured by training the estimator, using the first encoder and the second encoder, so that a result of estimating the characteristic of the material from at least one of the first feature vector and the second feature vector, calculated from the first data and the second data acquired, matches the correct information.

5 . The model generation method according to claim 1 , wherein
the first data indicates information regarding a local structure of a crystal of the material, and the second data indicates information regarding periodicity of a crystal structure of the material.

6 . The model generation method according to claim 5 , wherein the first data includes at least one of three-dimensional atomic position data, Raman spectroscopy data, nuclear magnetic resonance spectroscopy data, infrared spectroscopy data, mass spectrometry data, and X-ray absorption spectroscopy data.

7 . The model generation method according to claim 5 , wherein
the first data includes three-dimensional atomic position data, and a state of an atom in the material is expressed by at least one of a probability density function, a probability distribution function, and a probability mass function in the three-dimensional atomic position data.

8 . The model generation method according to claim 5 , wherein the second data includes at least one of X-ray diffraction data, neutron diffraction data, electron beam diffraction data, and total scattering data.

9 - 13 . (canceled)

14 . An estimation method comprising the steps of:
acquiring, by a computer, at least one of first data and second data regarding a crystal structure of a target material; converting, by the computer, at least one of the first data and second data acquired into at least one of a first feature vector and a second feature vector by using at least one of a trained first encoder and a trained second encoder; and estimating, by the computer, a characteristic of the target material from a value of at least one of the obtained first feature vector and second feature vector by using a trained estimator, wherein the second data indicates a property of a material with an index different from an index of the first data, a dimension of the first feature vector is the same as a dimension of the second feature vector, the trained first encoder and the trained second encoder are generated by machine learning using first data and second data for learning, the first data and the second data for learning include a positive sample and a negative sample, the positive sample includes a combination of first data and second data for the same material, the negative sample includes at least one of first data and second data for a material different from the material of the positive sample, machine learning for the first encoder and the second encoder is performed by training the first encoder and the second encoder so that values of a first feature vector and a second feature vector calculated from the first data and the second data of the positive sample are positioned close to each other, and a value of at least one of a first feature vector and a second feature vector calculated from at least one of the first data and the second data of the negative sample is positioned far from a value of at least one of the first feature vector and the second feature vector calculated from the positive sample, the trained estimator is generated by machine learning further using correct information indicating a characteristic of a material for learning, and the machine learning for the estimator is configured by training the estimator so that a result of estimating the characteristic of the material for learning from at least one of the first feature vector and the second feature vector, calculated from at least one of the first data and the second data for learning by using at least one of the first encoder and the second encoder, matches the correct information.

15 - 17 . (canceled)

18 . An estimation device comprising:
a target data acquisition unit configured to acquire at least one of first data and second data regarding a crystal structure of a target material; a conversion unit configured to convert at least one of the first data and second data acquired into at least one of a first feature vector and a second feature vector by using at least one of a trained first encoder and a trained second encoder; and an estimation unit configured to estimate a characteristic of the target material from a value of at least one of the obtained first feature vector and second feature vector by using a trained estimator, wherein the second data indicates a property of a material with an index different from an index of the first data, a dimension of the first feature vector is the same as a dimens

Assignees

• Omron Tateisi Electronics Co
• Inter Univ Research Institute Corporation High Energy Accelerator Research Organization

Inventors

• Taniai Tatsunori
• Ushiku Yoshitaka
• Chiba Naoya
• Suzuki Yuta
• Ono Kanta

Classifications

• G16C60/00

  Computational materials science, i.e. ICT specially adapted for investigating the physical or chemical properties of materials or phenomena associated with their design, synthesis, processing, characterisation or utilisation · CPC title

• G16C20/50

  Molecular design, e.g. of drugs · CPC title

• G16C20/70 Primary

  Machine learning, data mining or chemometrics · CPC title

• G06N3/045

  Combinations of networks · CPC title

• G06N3/08

  Learning methods · CPC title