In order to categorize AI mobile and edge applications into several classes regarding their security goals, a functional
evaluation was performed for numerous existing AI applications in this project. The evaluated applications originate
from a wide scope (biometry, smart home, smart living, smart devices, etc.), to identify a general classification that
suits a variety of applications. From this functional evaluation several classes were
Below, four different applications will be evaluated along with its determined classes, showing how a classification can
be achieved and to validate that the determined classes may be applied to numerous AI applications.
Where does the data come from?
Number of data classes?
One Class Classification / Novelty Detection
Type of data?
Where does training take place?
When is new training data added?
continuous (online learning)
Where is the model deployed?
What is known?
own histogramm-based novelty detector
What does the output look like?
ArcFace: Additive Angular Margin Loss for Deep Face Recognition