Options
Approximation of Neural Networks for Verification
Publikationstyp
Conference Paper
Date Issued
2019-04
Sprache
English
Author(s)
Institut
TORE-URI
Start Page
17
End Page
26
Citation
Methoden und Beschreibungssprachen zur Modellierung und Verifikation von Schaltungen und Systemen (MBMV 2019)
Scopus ID
ISBN of container
978-380074946-1
Statistical learning methods enable the adaptation of artificial neural networks (ANN) to complex problems. Meanwhile, formal properties can be verified on small ANNs under simplified assumptions. First we show a simple algorithm to convert neural networks into a system of equations with boundary conditions. In particular, we discuss how non-linear functions may be approximated. In experiments we study the impact of this approximation on the validity on the proof of formal guarantees.