US Patent Issued to International Business Machines on Aug. 12 for "Rules-based training of federated machine learning models" (Swedish, German Inventors)

ALEXANDRIA, Va., Aug. 12 -- United States Patent no. 12,387,158, issued on Aug. 12, was assigned to International Business Machines Corp. (Armonk, N.Y.).

"Rules-based training of federated machine learning models" was invented by Thomas Alexander Bertil Mattsson (Staffanstorp, Sweden) and Bilal Abdullah Abdullah Al-Saeedi (Munich).

According to the abstract* released by the U.S. Patent & Trademark Office: "Approaches presented herein enable training a federated machine learning model. More specifically, data is received from one or more sensors associated with an edge device. In response to the data exceeding a pre-determined threshold, a hazardous condition is identified. The hazardous condition is classified as valid or invalid, and in ...