US Patent Issued to MediaTek on Sept. 30 for "Compiling asymmetrically-quantized neural network models for deep learning acceleration" (Taiwanese Inventors)

ALEXANDRIA, Va., Sept. 30 -- United States Patent no. 12,430,542, issued on Sept. 30, was assigned to MediaTek Inc. (Hsinchu, Taiwan).

"Compiling asymmetrically-quantized neural network models for deep learning acceleration" was invented by Chih-Wen Goo (Hsinchu, Taiwan), Pei-Kuei Tsung (Hsinchu, Taiwan), Chih-Wei Chen (Hsinchu, Taiwan), Mingen Shih (Hsinchu, Taiwan), Shu-Hsin Chang (Hsinchu, Taiwan), Po-Hua Huang (Hsinchu, Taiwan), Ping-Yuan Tsai (Hsinchu, Taiwan), Shih-Wei Hsieh (Hsinchu, Taiwan) and You Yu Nian (Hsinchu, Taiwan).

According to the abstract* released by the U.S. Patent & Trademark Office: "A system receives a neural network model that includes asymmetric operations. Each asymmetric operation includes one or more fixed-po...