ALEXANDRIA, Va., Oct. 21 -- United States Patent no. 12,443,851, issued on Oct. 14, was assigned to GDM Holding LLC (Mountain View, Calif.).
"Augmenting attention-based neural networks to selectively attend to past inputs" was invented by Jack William Rae (London), Anna Potapenko (London) and Timothy Paul Lillicrap (London).
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 performing a machine learning task on a network input that is a sequence to generate a network output. In one aspect, one of the methods includes, for each particular sequence of layer inputs: for each attention layer in the neural network: m...