ALEXANDRIA, Va., March 3 -- United States Patent no. 12,566,958, issued on March 3, was assigned to Red Hat Inc. (Raleigh, N.C.).
"System and method of training a neural network" was invented by Mark Kurtz (Medford, Mass.) and Dan Alistarh (Meyrin, Switzerland).
According to the abstract* released by the U.S. Patent & Trademark Office: "A system and a method of training a Neural network (NN) model may include, receiving a pretrained NN model, that may include a plurality of layers, each associated with an activation matrix; selecting at least one, and performing an iterative training process on the layer. The iterative training process may include, applying an activation threshold to the activation matrix of the layer; measuring an accura...