US Patent Issued to INTERNATIONAL BUSINESS MACHINES on Oct. 14 for "Stepwise uncertainty-aware offline reinforcement learning under constraints" (Japanese Inventors)

ALEXANDRIA, Va., Oct. 21 -- United States Patent no. 12,443,678, issued on Oct. 14, was assigned to INTERNATIONAL BUSINESS MACHINES Corp. (Armonk, N.Y.).

"Stepwise uncertainty-aware offline reinforcement learning under constraints" was invented by Akifumi Wachi (Tokyo) and Takayuki Osogami (Yamato, Japan).

According to the abstract* released by the U.S. Patent & Trademark Office: "A computer-implemented method is provided for offline reinforcement learning with a dataset. The method includes training a neural network which inputs a state-action pair and outputs a respective Q function for each of a reward and one or more safety constraints, respectively. The neural network has a linear output layer and remaining non-linear layers being re...