Title:
LEARNING METHOD, INFORMATION PROCESSING SYSTEM, PROGRAM, AND LEARNING MODEL
Document Type and Number:
WIPO Patent Application WO/2024/101100
Kind Code:
A1
Abstract:
[Problem] To provide a learning method, an information processing system, a program, and a learning model with which it is possible to perform learning using data that is not explicitly given any feature complying with a law or the like. [Solution] According to one aspect of the present invention, there is provided a learning method that is performed by an information processing system. The learning method comprises an output step, an accepting step, and a learning step. In the output step, data that a learning model has generated during a learning process in accordance with a given condition, or without any condition given, is output. In the accepting step, an input of a result of determination of authenticity of the data by a determination means is accepted. In the learning step, learning by the learning model is allowed to proceed on the basis of the determination result.
More Like This:
Inventors:
YONEKURA KAZUO (JP)
WADA KAZUNARI (JP)
WADA KAZUNARI (JP)
Application Number:
PCT/JP2023/037717
Publication Date:
May 16, 2024
Filing Date:
October 18, 2023
Export Citation:
Assignee:
UNIV TOKYO (JP)
International Classes:
G06N20/00
Foreign References:
| JP2022072113A | 2022-05-17 | |||
| US20210271934A1 | 2021-09-02 | |||
| JP2020057172A | 2020-04-09 |
Other References:
SHU, DULE ET AL.: "3D Design Using Generative Adversarial Networks and Physics-Based Validation", ASME. J. MECH. DES, vol. 142, no. 7, 2020, pages 1 - 15, XP055787783, [retrieved on 20231204], DOI: https://doi.org/10.1115/1.4045419
Attorney, Agent or Firm:
IPX PATENT PARTNERS (JP)
Download PDF: