News Center

location: Home > News Center > News > Content

AIR Research|FedBCD: "Federal Average" Algorithm for Longitudinal Federal Learning

Source:       Time:2022-08-22

Recently, Institute for AI Industry Research, Tsinghua University (AIR), WeBank and the University of Minnesota have published a paper in the IEEE Transactions on Signal Processing (IEEE TSP, founded in 1991, is the top international journal in the discipline of "Information and Communication Engineering "). The research team has published a paper on FedBCD: A Communication-Efficient Collaborative Learning Framework for Distributed Features. The research team proposed an efficient communication-efficient collaborative learning framework for longitudinal federation learning of distributed features, which breaks through the traditional longitudinal federation learning communication bottleneck and improves security, and helps to circulate the value of data among cross-institutions.

For more details, please visit: AIR Research|FedBCD: "Federal Average" Algorithm for Longitudinal Federal Learning!

上一条:AIR DISCOVER|Dr. LIU Lingjie: Neural Expression and Rendering in 3D Real-world Scenes 下一条:Focusing on the Frontier, Exploring the Future: CICV 2022 Vehicle-Infrastructure Cooperated Autonomous Driving (VICAD) was successfully held


Address:12 / F, block C, Qidi science and technology building, Tsinghua Science and Technology Park, Haidian District, Beijing


Jing ICP Bei No. 15006448 | all rights reserved@ Institute of intelligent industry, Tsinghua University