Approximation of Neural Networks for Verification

Conference: MBMV 2019 - 22. Workshop „Methoden und Beschreibungssprachen zur Modellierung und Verifikation von Schaltungen und Systemen“
04/08/2019 - 04/08/2019 at Kaiserslautern, Deutschland

Proceedings: MBMV 2019

Pages: 10Language: englishTyp: PDF

Personal VDE Members are entitled to a 10% discount on this title

Bahnsen, Fin Hendrik; Fey, Goerschwin (Institute of Embedded Systems, Hamburg University of Technology, Hamburg, Germany)

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.