公开(公告)号：US20210287044A1

公开(公告)日：2021-09-16

申请号：US17104165

申请日：2020-11-25

Applicant: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY CO., LTD.

Inventor： Long LI ,  Haifeng WANG ,  Weibao GONG

IPC: G06K9/62 ,  G06F9/50 ,  G06N3/08

Abstract: A method for updating a parameter of a model, a distributed training system, and an electric device are related to a field of deep learning technologies. The method includes: obtaining a batch training period of batch training data to be trained for a model; increasing priorities of tasks ranked at the bottom in a sequence of gradient communication tasks for parameters of the model when the batch training period is greater than or equal to a preset period threshold; and performing a communication of gradients of the parameters and updating the parameters based on priorities of the gradient communication tasks for the parameters in the model.