US Patent Issued to NVIDIA on July 14 for "Quantized neural network training and inference" (Massachusetts, Texas, California Inventors)

ALEXANDRIA, Va., July 14 -- United States Patent no. 12,682,206, issued on July 14, was assigned to NVIDIA Corp. (Santa Clara, Calif.).

"Quantized neural network training and inference" was invented by Thierry Tambe (Cambridge, Mass.), Steve Dai (Union City, Calif.), Brucek Khailany (Austin, Texas) and Rangharajan Venkatesan (San Jose, Calif.).

According to the abstract* released by the U.S. Patent & Trademark Office: "One embodiment of a computer-implemented method for processing a neural network comprises receiving a first quantized matrix that corresponds to a portion of a multi-dimensional input tensor and has been quantized based on a first scale factor; and performing one or more computational operations using the first quantized ma...