US Patent Issued to INTERNATIONAL BUSINESS MACHINES on July 28 for "Concatenating input and output tensors for use in recurrent neural networks" (German, American Inventors)

ALEXANDRIA, Va., July 28 -- United States Patent no. 12,694,294, issued on July 28, was assigned to INTERNATIONAL BUSINESS MACHINES Corp. (Armonk, N.Y.).

"Concatenating input and output tensors for use in recurrent neural networks" was invented by Cedric Lichtenau (Stuttgart, Germany), Jonathan D. Bradbury (Poughkeepsie, N.Y.), Laith M. AlBarakat (Wappingers Falls, N.Y.) and Simon Weishaupt (Stuttgart, Germany).

According to the abstract* released by the U.S. Patent & Trademark Office: "A plurality of tensors is obtained, and the plurality of tensors is reformatted to provide a plurality of reformatted tensors of a select dimension. The reformatting includes adding padding to at least one reformatted tensor of the plurality of reformatted...