Identifying information assignment system, identifying information assignment method, and program therefor

Patent metadata

Field	Value
Publication number	US-10853571-B2
Application number	US-201716333657-A
Country	US
Kind code	B2
Filing date	Mar 13, 2017
Priority date	Oct 12, 2016
Publication date	Dec 1, 2020
Grant date	Dec 1, 2020

Abstract

An identifying information assignment system includes: a generating portion configured to generate, for a learning result obtained by attaining a predetermined capability through a predetermined learning process by machine learning, identifying information for identifying the predetermined learning process; and an assignment portion configured to assign the generated identifying information to the learning result.

First claim

The invention claimed is:

1. An identifying information assignment system, comprising: a processor configured with a program to perform operations comprising:
operation as a learning information acquiring portion configured to acquire learning request information comprising information to perform a machine learning process;
operation as a learning control portion configured to instruct a learning apparatus to perform the machine learning process, based on the learning request information;
operation as a factor information acquiring portion configured to acquire, from a factor information database, factors that may have an influence on a learning result obtained by the learning apparatus performing the machine learning process;
operation as an influence determining portion configured to determine whether or not each factor retrieved from the factor information database has an influence on the learning result, and to determine degrees of influence on the learning result, based on a correspondence relationship between the factors which influence the learning result according to conditions associated with the factors; and
operation as a generating portion configured to generate identifying information based on the factors determined to influence the learning result by the influence determining portion, wherein
operation as the learning control portion further comprises assigning the generated identifying information to the learning result.

2. The identifying information assignment system according to claim 1 , wherein the processor is configured with the program perform operations such that operation as the generating portion is further configured to:
generate, for a first learning result obtained by attaining a first capability through performing a first machine learning process, first identifying information for identifying the first machine learning process within a specific identification range; and
generate, for a second learning result obtained by attaining a second capability through performing a second machine learning process, second identifying information for identifying the second machine learning process such that the first machine learning process and the second machine learning process are distinguished from each other within the specific identification range.

3. The identifying information assignment system according to claim 2 , wherein the processor is configured with the program perform operations such that operation as the generating portion is further configured to generate the identifying information based on a factor that has an influence on a learning result within the specific identification range, among conditions that are associated with the first learning process and the second learning process being performed.

4. The identifying information assignment system according to claim 1 , wherein learning request information that is identifiable by the identifying information comprises at least one factor selected from among: a learning apparatus; a learning program; learning data; a number of times of learning; a time of learning; an object of learning; a method of learning; a learning requester; an acquisition source of learning data; and an acquisition method of learning data.

5. The identifying information assignment system according to claim 1 , wherein the processor is configured with the program perform operations further comprising: operation as a learning apparatus configured to perform the machine learning.

6. The identifying information assignment system according to claim 5 , wherein the processor is configured with the program perform operations such that operation as the learning apparatus is further configured to perform the machine learning according to learning request information to perform machine learning of a capability that is to be added to an object apparatus.

7. The identifying information assignment system according to claim 1 , wherein the processor is configured with the program perform operations such that operation as the generating portion is further configured to generate the identifying information, by encrypting, compressing, or listing a data sequence enumerating the factors determined to influence the learning result.

8. An identifying information assignment method for causing a computer to perform operations comprising:
acquiring learning request information comprising information to perform a machine learning process;
instructing a learning apparatus to perform the machine learning process, based on the learning request information;
acquiring, from a factor information database, factors that may have an influence on a learning result obtained by the learning apparatus performing the machine learning process;
determining whether or not each factor retrieved from the factor information database has an influence on the learning result, and degrees of influence on the learning result, based on a correspondence relationship between the factors which influence the learning result according to conditions associated with the factors;
generating identifying information based on the factors determined to influence the learning result; and
assigning the generated identifying information to the learning result.

9. A non-transitory computer-readable storage medium storing a program, which when read and executed, causes a computer to perform operations comprising:
acquiring learning request information comprising information to perform a machine learning process;
instructing a learning apparatus to perform the machine learning process, based on the learning request information;
acquiring, from a factor information database, factors that may have an influence on a learning result obtained by the learning apparatus performing the machine learning process;
determining whether or not each factor retrieved from the factor information database has an influence on the learning result, and degrees of influence on the learning result, based on a correspondence relationship between the factors which influence the learning result according to conditions associated with the factors;
generating identifying information based on the factors determined to influence the learning result; and
assigning the generated identifying information to the learning result.

10. The method according to claim 8 , further comprising:
generating, for a first learning result by attaining a first capability through performing a first machine learning process, first identifying information for identifying the first machine learning process within a specific identification range; and
generating, for a second learning result obtained by attaining a second capability through performing a second machine learning process, second identifying information for identifying the second machine learning process such that the first machine learning process and the second machine learning process are distinguished from each other within the specific identification range.

11. The method according to claim 10 , further comprising generating the identifying information based on a factor that has an influence on a learning result within the specific identification range, among conditions that are associated with the first learning process and the second learning process being performed.

12. The method according to claim 8 , wherein learning request information that is identifiable by the identifying information comprises at least one factor selected from among: a learning apparatus; a learning program; learning data; a number of times of learning; a time of learning; an object of learning; a method of learning; a learning requester; an acquisition source of learning data; and

Assignees

• Omron Tateisi Electronics Co

Inventors

• Ando Tanichi

Classifications

• G06N20/00 Primary

  Machine learning · CPC title

• G06F40/263 Primary

  Language identification · CPC title

• G06F40/279

  Recognition of textual entities · CPC title