Verification of Sigmoidal Artificial Neural Networks using iSAT

Autor:	Grundt, Dominik, Jurj, Sorin Liviu, Hagemann, Willem, Kröger, Paul, Fränzle, Martin
Rok vydání:	2022
Předmět:	Computer Science - Artificial Intelligence Computer Science - Machine Learning Computer Science - Symbolic Computation
Zdroj:	EPTCS 361, 2022, pp. 45-60
Druh dokumentu:	Working Paper
DOI:	10.4204/EPTCS.361.6
Popis:	This paper presents an approach for verifying the behaviour of nonlinear Artificial Neural Networks (ANNs) found in cyber-physical safety-critical systems. We implement a dedicated interval constraint propagator for the sigmoid function into the SMT solver iSAT and compare this approach with a compositional approach encoding the sigmoid function by basic arithmetic features available in iSAT and an approximating approach. Our experimental results show that the dedicated and the compositional approach clearly outperform the approximating approach. Throughout all our benchmarks, the dedicated approach showed an equal or better performance compared to the compositional approach. Comment: In Proceedings SNR 2021, arXiv:2207.04391
Databáze:	arXiv
Externí odkaz:	http://arxiv.org/abs/2207.06755 Zobrazit plný text záznamu View this record from Arxiv