US Patent Issued to GOOGLE on June 23 for "Fly parameter compression and decompression to facilitate forward and/or back propagation at clients during federated learning" (California Inventors)

ALEXANDRIA, Va., July 15 -- United States Patent no. 12,664,977, issued on June 23, was assigned to GOOGLE LLC (Mountain View, Calif.).

"Fly parameter compression and decompression to facilitate forward and/or back propagation at clients during federated learning" was invented by Tien-Ju Yang (Mountain View, Calif.), Yonghui Xiao (Sunnyvale, Calif.), Giovanni Motta (San Jose, Calif.), Francoise Beaufays (Mountain View, Calif.), Rajiv Mathews (Sunnyvale, Calif.) and Mingqing Chen (Saratoga, Calif.).

According to the abstract* released by the U.S. Patent & Trademark Office: "Implementations are directed to efficient federated learning of machine learning (ML) model(s) through on-the-fly decompression and compression of model parameters, of ...