US Patent Issued to Intel on Feb. 18 for "Sparsity-aware datastore for inference processing in deep neural network architectures" (California, Arizona Inventors)

ALEXANDRIA, Va., Feb. 19 -- United States Patent no. 12,229,673, issued on Feb. 18, was assigned to Intel Corp. (Santa Clara, Calif.).

"Sparsity-aware datastore for inference processing in deep neural network architectures" was invented by Deepak Mathaikutty (Chandler, Ariz.), Arnab Raha (Santa Clara, Calif.), Raymond Sung (San Francisco), Debabrata Mohapatra (Santa Clara, Calif.) and Cormac Brick (San Francisco).

According to the abstract* released by the U.S. Patent & Trademark Office: "Systems, apparatuses and methods may provide for technology that prefetches compressed data and a sparsity bitmap from a memory to store the compressed data in a decode buffer, where the compressed data is associated with a plurality of tensors, wherein ...