US Patent Issued to Cerebras Systems on May 27 for "Task synchronization for accelerated deep learning" (California Inventors)

ALEXANDRIA, Va., June 17 -- United States Patent no. 12,314,218, issued on May 27, was assigned to Cerebras Systems Inc. (Sunnyvale, Calif.).

"Task synchronization for accelerated deep learning" was invented by Sean Lie (Los Gatos, Calif.), Michael Morrison (Sunnyvale, Calif.), Srikanth Arekapudi (Los Altos Hills, Calif.), Michael Edwin James (San Carlos, Calif.) and Gary R. Lauterbach (Los Altos, Calif.).

According to the abstract* released by the U.S. Patent & Trademark Office: "Techniques in advanced deep learning provide improvements in one or more of accuracy, performance, and energy efficiency. An array of processing elements performs flow-based computations on wavelets of data. Each processing element has a compute element and a ro...