Title:
MACHINE LEARNING METHOD AND APPARATUS
Document Type and Number:
WIPO Patent Application WO/2023/033392
Kind Code:
A1
Abstract:
The present invention relates to a method for training a machine learning model and an apparatus therefor, the method comprising the steps of: obtaining first weight sets for medical machine learning models respectively, on the basis of medical data of each group, for a plurality of groups of medical data; evaluating performance of the medical machine learning models to which the respective first weight sets are applied, on the basis of a plurality of characteristic evaluation methods; selecting one best first weight set in which the performance is highly evaluated from among the plurality of first weight sets, for each characteristic evaluation method; and applying, to the medical machine learning models, a second weight set in which the plurality of first best weight sets are combined.
Inventors:
CHO KYUNG JAE (KR)
CHOI JAE WOO (KR)
SHIN YUN SEOB (KR)
TAE YUN WON (KR)
CHOI JAE WOO (KR)
SHIN YUN SEOB (KR)
TAE YUN WON (KR)
Application Number:
PCT/KR2022/011816
Publication Date:
March 09, 2023
Filing Date:
August 09, 2022
Export Citation:
Assignee:
VUNO INC (KR)
International Classes:
G06N20/00; G06N3/04; G06N3/08
Domestic Patent References:
| WO2021003142A1 | 2021-01-07 |
Foreign References:
| KR20210025108A | 2021-03-08 | |||
| KR102179090B1 | 2020-11-16 | |||
| KR20210068707A | 2021-06-10 | |||
| KR102052624B1 | 2019-12-05 |
Attorney, Agent or Firm:
KBK & ASSOCIATES (KR)
Download PDF: