公开(公告)号：US20210004718A1

公开(公告)日：2021-01-07

申请号：US16921207

申请日：2020-07-06

Applicant: Beijing Baidu Netcom Science and Technology Co., Ltd.

Inventor： Bing Ren ,  Shengwen Yang ,  Xuhui Zhou

IPC: G06N20/00 ,  H04L9/30 ,  H04L9/00

Abstract: A method and device for training a model based on federated learning are provided. The method includes: receiving a second original independent variable calculated value from a second data provider device; the second original independent variable calculated value being calculated by the second data provider device according to a second original independent variable and a second model parameter; calculating a dependent variable estimation value according to a first model parameter initial value of a first provider device, a first original independent variable of the first data provider device, and the second original independent variable calculated value; calculating a difference between a dependent variable of the first data provider device and the dependent variable estimation value; calculating a gradient of a loss function with respect to a first model parameter, according to the difference; and updating the first model parameter according to the gradient of the loss function with respect to the first model parameter.

公开(公告)号：US12056582B2

公开(公告)日：2024-08-06

申请号：US16921207

申请日：2020-07-06

Applicant: Beijing Baidu Netcom Science and Technology Co., Ltd.

Inventor： Bing Ren ,  Shengwen Yang ,  Xuhui Zhou

IPC: G06N20/00 ,  G06F21/60 ,  H04L9/00 ,  H04L9/08 ,  H04L9/30

CPC classification number: G06N20/00 ,  G06F21/602 ,  H04L9/008 ,  H04L9/0825 ,  H04L9/3073 ,  G06F2221/2107

Abstract: A method and device for training a model based on federated learning are provided. The method includes: receiving a second original independent variable calculated value from a second data provider device; the second original independent variable calculated value being calculated by the second data provider device according to a second original independent variable and a second model parameter; calculating a dependent variable estimation value according to a first model parameter initial value of a first provider device, a first original independent variable of the first data provider device, and the second original independent variable calculated value; calculating a difference between a dependent variable of the first data provider device and the dependent variable estimation value; calculating a gradient of a loss function with respect to a first model parameter, according to the difference; and updating the first model parameter according to the gradient of the loss function with respect to the first model parameter.