US Patent Issued to Moffett International on July 7 for "Processor, method, and system for accelerating tensor transpose for machine learning" (California Inventors)

ALEXANDRIA, Va., July 7 -- United States Patent no. 12,675,293, issued on July 7, was assigned to Moffett International Co. Ltd. (Kowloon, Hong Kong).

"Processor, method, and system for accelerating tensor transpose for machine learning" was invented by Changxu Zhang (Los Altos, Calif.), Zhibin Xiao (Los Altos, Calif.), Enxu Yan (Los Altos, Calif.), Xiaoqian Zhang (Los Altos, Calif.) and Renjie Chen (Los Altos, Calif.).

According to the abstract* released by the U.S. Patent & Trademark Office: "The present invention relates to a processor designed to optimize memory bandwidth utilization for tensor transpositions in machine learning. It features an instruction decoder that decodes tensor transpose instructions for an input tensor, disting...