ALEXANDRIA, Va., March 19 -- United States Patent no. 12,254,412, issued on March 18, was assigned to Snap Inc. (Santa Monica, Calif.).
"Optimizer based prunner for neural networks" was invented by Enxu Yan (Los Altos, Calif.), Sergey Tulyakov (Marina del Rey, Calif.), Aleksei Podkin (Santa Monica, Calif.) and Aleksei Stoliar (Marina del Rey, Calif.).
According to the abstract* released by the U.S. Patent & Trademark Office: "A neural network pruning system can sparsely prune neural network models using an optimizer based approach that is agnostic to the model architecture being pruned. The neural network pruning system can prune by operating on the parameter vector of the full model and the gradient vector of the loss function with respe...