Volkswagen Ag

Dspace Gmbh

Neurocat Gmbh

Florens Fabian Gressner

Peter Schlicht

Fabian Hüger

Nico Maurice Schmidt

Stephan Hinze

Frank Kretschmer

Felix Assion

Feix Assion

The invention relates to a method for generating disturbed input data for a neural network for analyzing sensor data, in particular digital images, of a driver assistance system, in which a first metric is defined which indicates how the magnitude of a change in sensor data is measured, a second metric is defined which indicates where a disturbance of sensor data is directed, an optimization problem is generated from a combination of the first metric and second metric, the optimization problem is solved by means of at least one solution algorithm, wherein the solution indicates a target disturbance of the input data, and disturbed input data is generated from sensor data for the neural network by means of the target disturbance.

Method and generator for generating disturbed input data for a neural network

Adversarial attacks pose a threat to the functioning of AI systems. A defense generator for dynamically generating at least one AI defense module is provided. A core feature of the invention is that a distribution function is determined for model data. The invention is based on the assumption that the model data belong to a model manifold, or have a similar statistical behavior. Thus, it is possible to determine for an input data set whether the data of the input data set can be associated with an adversarial attack. This is the case, for example, if statistical anomalies are found in the input data set.

Defense generator, method for preventing an attack on an AI unit, and computer-readable storage medium

Growing community of inventors

Berlin, Germany

Florens Fabian Gressner