Patent · US Active

Method, apparatus and computer program for generating robust automated learning systems and testing trained automated learning systems

US11386328B2 · kind B2 · utility

1Cited by

1References

26Claims

0Family size

Assignees

Robert Bosch GmbH · DE
CARNEGIE MELLON UNIVERSITY · US

Inventors

Eric Wong · Lexington, US
Frank Schmidt · Leonberg, DE
Jan Hendrick Metzen · Hessenhöfe, DE
Jeremy Zico Kolter · Pittsburgh, US

Key dates

Filing date	Oct 29, 2018
Grant date	Jul 12, 2022
Priority date	—
Expiry date	Dec 6, 2040

Classification

Technology area (CPC G)Physics
CPC primaryG06N3/088
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

In a method for training a first neural network a superposed classification is back-propagated through a second neural network. An output value of the second neural network is utilized to determine whether the input of the first neural network is adversarial.

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