ALEXANDRIA, Va., Feb. 5 -- United States Patent no. 12,217,147, issued on Feb. 4, was assigned to Cerebras Systems Inc. (Sunnyvale, Calif.).
"Advanced wavelet filtering for accelerated deep learning" was invented by Michael Morrison (Sunnyvale, Calif.), Michael Edwin James (San Carlos, Calif.), Sean Lie (Los Altos, Calif.), Srikanth Arekapudi (Los Altos Hills, Calif.) and Gary R. Lauterbach (Los Altos, Calif.).
According to the abstract* released by the U.S. Patent & Trademark Office: "Techniques in wavelet filtering for advanced deep learning provide improvements in one or more of accuracy, performance, and energy efficiency. An array of processing elements comprising a portion of a neural network accelerator performs flow-based computat...