US Patent Issued to DeepMind Technologies on July 22 for "Neural network optimization using curvature estimates based on recent gradients" (Canadian, British Inventors)

ALEXANDRIA, Va., July 23 -- United States Patent no. 12,367,387, issued on July 22, was assigned to DeepMind Technologies Ltd. (London).

"Neural network optimization using curvature estimates based on recent gradients" was invented by David William Saxton (London) and Eshaan Nichani (San Diego, Canada).

According to the abstract* released by the U.S. Patent & Trademark Office: "Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for training a neural network by estimating the objective function curvature based on current and previous gradients. In one aspect, a method comprises: sampling a batch of training data; and for each neural network parameter: determining, based on the current batch o...