Patent · US Active

Pruning and retraining method for a convolution neural network

US11200495B2 · kind B2 · utility

2Cited by

1References

13Claims

0Family size

Assignee

Vivante Corporation · US

Inventors

Xin Wang · Geelong, AU
Shang-Hung Lin · San Jose, US

Key dates

Filing date	Sep 8, 2017
Grant date	Dec 14, 2021
Priority date	—
Expiry date	Jul 21, 2040

Classification

Technology area (CPC G)Physics
CPC primaryG06N3/09
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

A convolution neural network (CNN) model is trained and pruned at a pruning ratio. The model is then trained and pruned one or more times without constraining the model according to any previous pruning step. The pruning ratio may be increased at each iteration until a pruning target is reached. The model may then be trained again with pruned connections masked. The process of pruning, retraining, and adjusting the pruning ratio may also be repeated one or more times with a different pruning target.

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