US Patent Issued to SEOUL NATIONAL UNIVERSITY R&DB FOUNDATION on Sept. 15 for "Algorithm and method for dynamically varying quantization precision of deep learning network" (South Korean Inventors)

ALEXANDRIA, Va., Sept. 15 -- United States Patent no. 12,738,049, issued on Sept. 15, was assigned to SEOUL NATIONAL UNIVERSITY R&DB FOUNDATION (Seoul, South Korea).

"Algorithm and method for dynamically varying quantization precision of deep learning network" was invented by Ook Sang Yoo (Gimpo-si, South Korea), Hyuk Jae Lee (Seongnam-si, South Korea), Soo Jung Ryu (Hwaseong-si, South Korea), Ji Yea Chon (Seoul, South Korea) and Kyeong Jong Lim (Seongnam-si, South Korea).

According to the abstract* released by the U.S. Patent & Trademark Office: "An image recognition method includes the steps of: for a deep learning network that carries out object recognition on a random image, carrying out quantization corresponding to the number of a p...