FEDERATED LEARNING METHOD, DEVICE, EQUIPMENT, AND STORAGE MEDIUM

WIPO Patent Application WO/2021/115480

A1

A federated learning method, related to the field of artificial intelligence, comprising: acquiring task configuration information of at least two clients participating in a target federated learning task; the task configuration information being obtained by pooling task configuration data received by the clients; initializing model training configuration information in the federated learning task on the basis of the task configuration information; executing a model training operation of the target federated learning task on the basis of the initialized model training configuration information, transmitting a model update parameter generated when executing the model training operation to the corresponding clients, thus allowing the clients to update a local federated learning model, and obtaining trained federated learning models corresponding to the clients. Also related to the field of machine learning.

JP2022006853	INFORMATION PROCESSING SYSTEM AND COMPRESSION CONTROL METHOD
JP7286259	Feature vector generation device, feature vector generation method, and feature vector generation program
JP7268731	LEARNING DATA GENERATION DEVICE, LEARNING DEVICE, LEARNING DATA GENERATION METHOD, AND LEARNING DATA GENERATION PROGRAM

ZHOU XUELI (CN)
ZHU ENDONG (CN)
ZHANG QIAN (CN)
LING HAITING (CN)

PCT/CN2020/136112

June 17, 2021

December 14, 2020

Click for automatic bibliography generation  

PING AN TECH SHENZHEN CO LTD (CN)

G06N20/00

CN111768008A	2020-10-13
CN107644255A	2018-01-30
CN110428058A	2019-11-08
CN109144729A	2019-01-04

ADVANCE CHINA IP LAW OFFICE (CN)

View/Download PDF      PDF Help

Previous Patent: DISPLAY CONTROL METHOD AND ELECTRONIC DEVICE

Next Patent: TERMINAL CONTROL METHOD AND DEVICE, TERMINAL AND STORAGE MEDIUM