US Patent Issued to Robert Bosch on Jan. 28 for "Method and system for breaking backdoored classifiers through adversarial examples" (Pennsylvania Inventors)

ALEXANDRIA, Va., Jan. 29 -- United States Patent no. 12,210,619, issued on Jan. 28, was assigned to Robert Bosch GmbH (Germany).

"Method and system for breaking backdoored classifiers through adversarial examples" was invented by Mingjie Sun (Pittsburgh), Jeremy Kolter (Pittsburgh) and Filipe J. Cabrita Condessa (Pittsburgh).

According to the abstract* released by the U.S. Patent & Trademark Office: "A computer-implemented method for training a machine learning network includes receiving an input data from one or more sensors, selecting one or more batch samples from the input data, wherein the batch samples include one or more perturbed samples from a source class configured to be misclassified into a target class, identifying the one or...