US Patent Issued to Uber Technologies on March 3 for "Training adaptable neural networks based on evolvability search" (California, New York Inventors)

ALEXANDRIA, Va., March 3 -- United States Patent no. 12,566,966, issued on March 3, was assigned to Uber Technologies Inc. (San Francisco).

"Training adaptable neural networks based on evolvability search" was invented by Alexander P. Gajewski (New York), Jeffrey Michael Clune (San Francisco), Kenneth Owen Stanley (San Francisco) and Joel Anthony Lehman (San Francisco).

According to the abstract* released by the U.S. Patent & Trademark Office: "Systems and methods are disclosed herein for training neural networks that can be adapted to new inputs, new tasks, new environment, etc. by re-training them efficiently. A parameter vector is initialized for a neural network. Perturbed parameter vectors are determined using the parameter vector. B...