US Patent Issued to NVIDIA on Feb. 17 for "Optimizing parameter estimation for training neural networks" (Maryland, California, Idaho, New Jersey, Tennessee Inventors)

ALEXANDRIA, Va., Feb. 17 -- United States Patent no. 12,554,796, issued on Feb. 17, was assigned to NVIDIA Corp. (Santa Clara, Calif.).

"Optimizing parameter estimation for training neural networks" was invented by Daguang Xu (Potomac, Md.), Holger Reinhard Roth (Rockville, Md.), Dong Yang (Pocatello, Idaho), Andriy Myronenko (San Francisco), Ali Hatamizadeh (Los Angeles), Vishwesh Nath (Nashville, Tenn.) and Anas Zainul Abidin (Fort Lee, N.J.).

According to the abstract* released by the U.S. Patent & Trademark Office: "Apparatuses, systems, and techniques estimate parameters to train one or more neural networks based on uniqueuss of training data. In at least one embodiment, a subset of training data is selected and used to estimate para...