Title:
VALIDATING PERFORMANCE OF A NEURAL NETWORK TRAINED USING LABELED TRAINING DATA
Document Type and Number:
WIPO Patent Application WO/2021/033872
Kind Code:
A1
Abstract:
A method for validating performance of a neural network trained using labeled training and validation data is provided. The method includes: determining proposed model parameters as potential updates to the neural network using the labeled validation data, performing a short-term validation on the proposed model parameters applied to the neural network based on the labeled validation data by comparing a first performance output based on the proposed model parameters and a second performance output based on currently-existing model parameters applied to the neural network, updating the currently-existing model parameters with the proposed model parameters when the second performance output outperforms the first performance output with respect to the labeled validation data, performing a long-term validation on the updated currently-existing model parameters applied to the neural network, and performing an operation when a difference between the original model parameters and the updated currently-existing model parameters lies within a threshold.
Inventors:
JUNG NAMSOON (KR)
Application Number:
PCT/KR2020/006249
Publication Date:
February 25, 2021
Filing Date:
May 13, 2020
Export Citation:
Assignee:
LG ELECTRONICS INC (KR)
International Classes:
G06N3/08; G06N3/04; G06N20/00
Foreign References:
US20170124486A1 | 2017-05-04 | |||
US20150356457A1 | 2015-12-10 | |||
US20180300576A1 | 2018-10-18 | |||
US20170024641A1 | 2017-01-26 | |||
KR20170134508A | 2017-12-06 |
Attorney, Agent or Firm:
HAW, Yong Noke (KR)
Download PDF:
Previous Patent: SOFT CAPSULE CONTAINING TYNDALIZED BACTERIA IN FILM
Next Patent: GENERATING TRAINING AND VALIDATION DATA FOR MACHINE LEARNING
Next Patent: GENERATING TRAINING AND VALIDATION DATA FOR MACHINE LEARNING