Title:
MACHINE LEARNING DEVICE, INFERENCE DEVICE, AND TRAINED MODEL
Document Type and Number:
WIPO Patent Application WO/2023/047713
Kind Code:
A1
Abstract:
This machine learning device 100 comprises: a first main component analysis unit 10 for analyzing training data with regard to main components, thereby generating two or more main components of the training data; a first image data generation unit 20 for allocating a first main component among the two or more main components to the X coordinate of an XY plane and allocating a second main component to the Y coordinate of the XY plane, thereby generating virtual image data; and a training unit 30 for performing machine learning using the image data as input data, thereby generating a trained model.
Inventors:
OKUSHIRO HIDETAKA (JP)
Application Number:
PCT/JP2022/022868
Publication Date:
March 30, 2023
Filing Date:
June 07, 2022
Export Citation:
Assignee:
JVCKENWOOD CORP (JP)
International Classes:
G06N20/00; G06T7/00
Foreign References:
| JP2019066993A | 2019-04-25 | |||
| US20200218806A1 | 2020-07-09 | |||
| JP2018124639A | 2018-08-09 | |||
| JPH06174689B1 |
Other References:
YUTO KANAZAWA , TAKASHI NAGAOKA , YUICHI KIMURA , MITSUTAKA NEMOTO: "[Short Paper] Automatic diagnosis of melanoma by deep learning - Dimensional reduction by principal component analysis of HSD", IEICE TECHNICAL REPORT, vol. 120, no. 431 (MI2020-82), 8 March 2021 (2021-03-08), JP , pages 150 - 151, XP009544889, ISSN: 2432-6380
Attorney, Agent or Firm:
MORISHITA Sakaki (JP)
Download PDF: