ALEXANDRIA, Va., June 25 -- United States Patent no. 12,340,313, issued on June 24, was assigned to MOFFETT INTERNATIONAL Co. Ltd. (Hong Kong).

"Neural network pruning method and system via layerwise analysis" was invented by Enxu Yan (Los Altos, Calif.), Dongkuan Xu (Los Altos, Calif.) and Jiachao Liu (Los Altos, Calif.).

According to the abstract* released by the U.S. Patent & Trademark Office: "Embodiments disclosed herein allowed neural networks to be pruned. The inputs and outputs generated by a reference neural network are used to prune the reference neural network. The pruned neural network may have a subset of the weights that are in the reference neural network."

The patent was filed on Nov. 30, 2020, under Application No. 17/10...