Patent · US Active

Detection and prevention of adversarial deep learning

US11526601B2 · kind B2 · utility

1Cited by

4References

20Claims

0Family size

Assignee

The Regents of the University of California · US

Inventors

Bita Darvish Rouhani · San Diego, US
Tara Javidi · Ann Arbor, US
Farinaz Koushanfar · Houston, US
Mohammad Samragh Razlighi · San Diego, US

Key dates

Filing date	Jul 12, 2018
Grant date	Dec 13, 2022
Priority date	—
Expiry date	Jul 18, 2039

Classification

Technology area (CPC G)Physics
CPC primaryG06F2221/034
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

A method for detecting and/or preventing an adversarial attack against a target machine learning model may be provided. The method may include training, based at least on training data, a defender machine learning model to enable the defender machine learning model to identify malicious input samples. The trained defender machine learning model may be deployed at the target machine learning model. The trained defender machine learning model may be coupled with the target machine learning model to at least determine whether an input sample received at the target machine learning model is a malicious input sample and/or a legitimate input sample. Related systems and articles of manufacture, including computer program products, are also provided.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.