US Patent Issued to NEC on Aug. 26 for "Self-supervised multimodal representation learning with cascade positive example mining" (California, Wisconsin Inventors)

ALEXANDRIA, Va., Aug. 26 -- United States Patent no. 12,400,449, issued on Aug. 26, was assigned to NEC Corp. (Tokyo).

"Self-supervised multimodal representation learning with cascade positive example mining" was invented by Farley Lai (Santa Clara, Calif.), Asim Kadav (Mountain View, Calif.) and Cheng-En Wu (Madison, Wis.).

According to the abstract* released by the U.S. Patent & Trademark Office: "A method for model training and deployment includes training, by a processor, a model to learn video representations with a self-supervised contrastive loss by performing progressive training in phases with an incremental number of positive instances from one or more video sequences, resetting the learning rate schedule in each of the phases, ...