Title:
MACHINE LEARNING DEVICE, LEARNING MODEL GENERATION METHOD, AND PROGRAM
Document Type and Number:
WIPO Patent Application WO/2021/193931
Kind Code:
A1
Abstract:
The present invention improves accuracy in machine learning and reduces running cost by, for example, performing update so as to obtain a new learning model generated using selected data. In the present invention, a data collection unit collects data for executing machine learning. The data storage unit stores the collected data therein. A data selection unit selects data for updating the existing learning model for use in machine learning among data stored in the data storage unit. Further, a learning model generation unit generates a new learning model through machine learning on the basis of the selected data, and an update unit at least updates the existing learning model so as to obtain a new learning model generated by the learning model generation unit.
Inventors:
KIKUCHI TAKAYOSHI (JP)
OHUCHIDA SHUN (JP)
OHUCHIDA SHUN (JP)
Application Number:
PCT/JP2021/012928
Publication Date:
September 30, 2021
Filing Date:
March 26, 2021
Export Citation:
Assignee:
KYB CORP (JP)
International Classes:
G06N20/00
Domestic Patent References:
| WO2019176354A1 | 2019-09-19 |
Foreign References:
| US20190378044A1 | 2019-12-12 | |||
| CN109214559A | 2019-01-15 | |||
| JP2020024633A | 2020-02-13 |
Other References:
YANGQING JIAEVAN SHELHAMERJEFF DONAHUESERGEY KARAYEVJONATHAN LONGROSS GIRSHICKSERGIO GUADARRAMATREVOR DARRELL: "Proceedings of the 22nd ACM international conference on Multimedia", ACM, article "Caffe: Convolutional architecture for fast feature embedding", pages: 675 - 678
See also references of EP 4131090A4
See also references of EP 4131090A4
Attorney, Agent or Firm:
KATO, Seiji (JP)
Download PDF:
Previous Patent: HOLLOW FIBER MEMBRANE MODULE
Next Patent: ELECTROSTATIC CAPACITANCE SENSOR, SENSOR UNIT, AND ROBOT
Next Patent: ELECTROSTATIC CAPACITANCE SENSOR, SENSOR UNIT, AND ROBOT