US Patent Issued to Intel on Feb. 17 for "System and method for compressing convolutional neural networks" (Oregon Inventor)

ALEXANDRIA, Va., Feb. 17 -- United States Patent no. 12,554,988, issued on Feb. 17, was assigned to Intel Corp. (Santa Clara, Calif.).

"System and method for compressing convolutional neural networks" was invented by Anthony Daniel Rhodes (Portland, Ore.).

According to the abstract* released by the U.S. Patent & Trademark Office: "An apparatus is provided to compress CNN models using a combination of filter pruning and tensor decomposition. For example, the apparatus accesses a trained CNN that includes convolutional tensors. The apparatus prunes the filters of a convolutional tensor to generate a sparse tensor. Further, the apparatus decomposes the sparse tensor to generate a low-rank approximation of the sparce tensor. The low-rank appr...