US Patent Issued to Microsoft Technology Licensing on Dec. 30 for "Resource-efficient training of a sequence-tagging model" (California Inventors)

ALEXANDRIA, Va., Dec. 31 -- United States Patent no. 12,511,548, issued on Dec. 30, was assigned to Microsoft Technology Licensing LLC (Redmond, Wash.).

"Resource-efficient training of a sequence-tagging model" was invented by Wen Cui (Los Altos, Calif.), Keng-hao Chang (San Jose, Calif.), Pai Chun Lin (Fremont, Calif.), Mohammadreza Khalilishoja (Sunnyvale, Calif.) and Eren Manavoglu (Menlo Park, Calif.).

According to the abstract* released by the U.S. Patent & Trademark Office: "A technique iteratively updates model weights of a teacher model and a student model. In operation, the teacher model produces noisy original pseudo-labeled training examples from unlabeled training examples. The technique weights the original pseudo-labeled tra...