US Patent Issued to GOOGLE on Jan. 28 for "Training a policy model for a robotic task, using reinforcement learning and utilizing data that is based on episodes, of the robotic task, guided by an engineered policy" (California Inventors)

ALEXANDRIA, Va., Jan. 29 -- United States Patent no. 12,210,943, issued on Jan. 28, was assigned to GOOGLE LLC (Mountain View, Calif.).

"Training a policy model for a robotic task, using reinforcement learning and utilizing data that is based on episodes, of the robotic task, guided by an engineered policy" was invented by Adrian Li (San Francisco), Benjamin Holson (Sunnyvale, Calif.), Alexander Herzog (San Jose, Calif.) and Mrinal Kalakrishnan (Mountain View, Calif.).

According to the abstract* released by the U.S. Patent & Trademark Office: "Implementations disclosed herein relate to utilizing at least one existing manually engineered policy, for a robotic task, in training an RL policy model that can be used to at least selectively rep...