Title:
NEURAL NETWORK TRAINING METHOD AND APPARATUS, AND ELECTRONIC DEVICE
Document Type and Number:
WIPO Patent Application WO/2020/143225
Kind Code:
A1
Abstract:
Disclosed are a neural network training method and apparatus, and an electronic device. The neural network training method comprises: inputting training data into a trained first neural network and a second neural network to be trained; determining a first feature map output by a preset layer of the first neural network and a second feature map output by the second neural network at the preset layer; determining a first loss function value of the second neural network on the basis of the first feature map and the second feature map; updating parameters of the second neural network on the basis of the first loss function value and a second loss function value of the second neural network; and taking the updated parameters of the second neural network as initial parameters of the second neural network to be trained, updating the parameters of the second neural network in an iterative manner, and if the updated second neural network meets a preset condition, obtaining a final trained second neural network. In this way, the precision of the trained second neural network is improved.
Inventors:
ZHOU HELONG (CN)
ZHANG QIAN (CN)
HUANG CHANG (CN)
ZHANG QIAN (CN)
HUANG CHANG (CN)
Application Number:
PCT/CN2019/100983
Publication Date:
July 16, 2020
Filing Date:
August 16, 2019
Export Citation:
Assignee:
NANJING INST OF ADVANCED ARTIFICIAL INTELLIGENCE LTD (CN)
International Classes:
G06N3/04
Foreign References:
| CN107247989A | 2017-10-13 | |||
| CN108664893A | 2018-10-16 | |||
| CN108764462A | 2018-11-06 | |||
| CN108960407A | 2018-12-07 | |||
| US20180268292A1 | 2018-09-20 |
Attorney, Agent or Firm:
NANJING XINHUIHENGCHENG INTELLECTUAL PROPERTY AGENT CO., LTD (CN)
Download PDF: