# US 2025/0095128 A1 · Learning method, learning apparatus, learning program, control method, control apparatus, and control program

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2025/0095128 A1
- Document ID: US20250095128A1
- Status: Published application
- Kind: A1
- Application number: 18/815,527
- Filed: 2024-08-26
- Published: 2025-03-20
- Earliest claimed: 2022-02-24
- Assignees: Riken
- Applicants: Riken
- Inventors: Toru UJIHARA; Masaru ISONO; Masaki TAKAISHI; Yosuke TSUNOOKA; Ichiro TAKEUCHI; Kentaro KUTSUKAKE
- Web page: https://patentgrep.com/patent/US20250095128A1

## CPC

- G06T 7/0004: {Industrial image inspection}
- G06N 3/0455: Auto-encoder networks; Encoder-decoder networks
- G06N 20/00: Machine learning
- G06Q 50/10: Services
- G06T 2207/10024: Color image
- G06T 2207/20081: Training; Learning
- G06T 2207/30128: Food products
- G06T 2207/30148: Semiconductor; IC; Wafer
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06Q: INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR

## Abstract

A learning method includes: causing an observer to perceive information to be compared by the observer from among a plurality of pieces of perceivable information; acquiring evaluation for the information made by the observer; and rating the plurality of pieces of information based on the evaluation made by the observer for the plurality of pieces of information acquired by a plurality of repetitions of the causing the observer to perceive the information and the acquiring the evaluation.

## Claims

### Claim 1

A learning method comprising:
causing an observer to perceive a plurality of pieces of perceivable information by pairwise comparison;
acquiring evaluation for the information made by the observer;
rating the plurality of pieces of information by a plurality of repetitions of the causing the observer to perceive the information and the acquiring the evaluation;
generating a latent space by dimensional compression of the plurality of pieces of information; and
searching for a latent variable that matches a predetermined condition in the latent space.

### Claim 2 (depends on 1)

The learning method according to claim 1, wherein
the rating the plurality of pieces of information includes rating the plurality of pieces of information by Elo rating.

### Claim 3 (depends on 1)

The learning method according to claim 1, wherein
the generating the latent space includes generating a latent space by dimensional compression of the plurality of pieces of information such that similar information is closely placed.

### Claim 4 (depends on 3)

The learning method according to claim 3, wherein
the generating the latent space includes generating the latent space by a variational autoencoder using a score assigned to each of the plurality of pieces of information in the rating.

### Claim 5 (depends on 3)

The learning method according to claim 3, wherein
the searching includes searching for a latent variable with the largest score.

### Claim 6 (depends on 1)

The learning method according to claim 1, wherein
the information represents an image or a moving image.

### Claim 7 (depends on 6)

The learning method according to claim 6, wherein
the information represents an image related to the production of a material.

### Claim 8 (depends on 7)

The learning method according to claim 7, wherein
the information represents an image visualizing temperature and flow distributions of a raw material for producing the material.

### Claim 9 (depends on 8)

The learning method according to claim 8, wherein
the material is a SiC crystal.

### Claim 10

A learning apparatus comprising circuitry configured to:
cause an observer to perceive a plurality of pieces of perceivable information by pairwise comparison;
acquire evaluation for the information made by the observer;
rate the plurality of pieces of information by a plurality of repetitions of the providing the information and the acquiring the evaluation;
generate a latent space by dimensional compression of the plurality of pieces of information; and
search for a latent variable that matches a predetermined condition in the latent space.

### Claim 11

A control method comprising:
causing an observer by pairwise comparison to perceive information that expresses the state of a control target;
acquiring evaluation made by the observer for information expressing the state of the control target at a first time point and information expressing the state of the control target at a second time point different from the first time point; and
controlling the control target in accordance with a control parameter corresponding to a latent variable that matches a predetermined condition searched for in a latent space generated by dimensional compression of a plurality of pieces of information, the plurality of pieces of information being rated based on the evaluation.

### Claim 12 (depends on 11)

The control method according to claim 11, comprising:
rating the state of the control target based on the evaluation, wherein
the controlling includes controlling the control target based on the rating.

### Claim 13 (depends on 11)

The control method according to claim 11, wherein
the information represents an image or a moving image.

### Claim 14 (depends on 13)

The control method according to claim 13, wherein
the control target is an apparatus for producing a material, and
the information represents an image related to the production of the material.

### Claim 15 (depends on 14)

The control method according to claim 14, wherein
the information represents an image visualizing temperature and flow distributions of a raw material for producing the material.

### Claim 16 (depends on 15)

The control method according to claim 15, wherein
the material is a SiC crystal, and
the controlling is to control at least one of the rotation speed of a seed crystal, the rotation speed of a crucible, the relative position of the crucible and a coil, the position of the coil, the level of a high-temperature liquid metal, the level of a meniscus of the high-temperature liquid metal, and a set temperature.

### Claim 17

A control apparatus comprising circuitry configured to:
cause an observer by pairwise comparison to perceive information that expresses the state of a control target;
acquire evaluation made by the observer for information expressing the state of the control target at a first time point and information expressing the state of the control target at a second time point different from the first time point; and
control the control target in accordance with a control parameter corresponding to a latent variable that matches a predetermined condition searched for in a latent space generated by dimensional compression of a plurality of pieces of information, the plurality of pieces of information being rated based on the evaluation.

### Claim 18

A non-transitory recording medium having embodied thereon a learning program comprising computer-implemented modules including:
a module that causes an observer to perceive a plurality of pieces of perceivable information by pairwise comparison;
a module that acquires evaluation for the information made by the observer;
a module that rates the plurality of pieces of information by a plurality of repetitions of the providing the information and the acquiring the evaluation;
a module that generates a latent space by dimensional compression of the plurality of pieces of information; and
a module that searches for a latent variable that matches a predetermined condition in the latent space.

### Claim 19

A non-transitory recording medium having embodied thereon a control program comprising computer-implemented modules including:
a module that causes an observer by pairwise comparison to perceive information that expresses the state of a control target;
a module that acquires evaluation made by the observer for information expressing the state of the control target at a first time point and information expressing the state of the control target at a second time point different from the first time point; and
a module that controls the control target in accordance with a control parameter corresponding to a latent variable that matches a predetermined condition searched for in a latent space generated by dimensional compression of a plurality of pieces of information, the plurality of pieces of information being rated based on the evaluation.