US Patent Issued to Verizon Patent and Licensing on Sept. 22 for "Systems and methods for providing a fraud model for proactively preventing attacks on machine learning models" (Indian Inventors)

ALEXANDRIA, Va., Sept. 24 -- United States Patent no. 12,743,505, issued on Sept. 22, was assigned to Verizon Patent and Licensing Inc. (Basking Ridge, N.J.).

"Systems and methods for providing a fraud model for proactively preventing attacks on machine learning models" was invented by Srilakshmi Jayaraman (Chennai, India), Aiswarya Devarajan (Chennai, India) and Padma Rangarajan (Chennai, India).

According to the abstract* released by the U.S. Patent & Trademark Office: "A device may perform adversarial training of a fraud model based on a primary model and with different training data than used to train the primary model, and may receive, from a user, a request for the primary model to generate a prediction. The device may cause the req...