Patent · US Active

Method for updating neural network and electronic device

US11328180B2 · kind B2 · utility

0Cited by

0References

20Claims

0Family size

Assignee

Beijing Horizon Robotics Technology Research and Development Co., Ltd. · CN

Inventors

Yonghao Xu · Hubei, CN
Qian Zhang · Cypress, US
Guoli Wang · Beijing, CN
Chang Huang · Beijing, CN

Key dates

Filing date	Oct 29, 2019
Grant date	May 10, 2022
Priority date	—
Expiry date	May 15, 2040

Classification

Technology area (CPC G)Physics
CPC primaryG06N3/048
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

Disclosed are a method for updating a neural network and an electronic device. The method includes: inputting a first image set having tag information into a first depth neural network, and determining a cross entropy loss value of the first image set by using the first depth neural network; inputting a second image set having no tag information separately into the first depth neural network and a second depth neural network, and determining a consistency loss value of the second image set, the first depth neural network and the second depth neural network having the same network structure; updating parameters of the first depth neural network based on the cross entropy loss value and the consistency loss value; and updating parameters of the second depth neural network based on the updated parameters of the first depth neural network.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.