Formal verification of deep neural networks

• Published in: PROBLEMS IN PROGRAMMING
• Date: 2024
• Issue: 2-3
• Pages: 253-262
• ISSN: 1727-4907
• Author Affiliations: B.O. Panchuk — Glushkov Institute of Cybernetics NAS of Ukraine
• Keywords: штучна нейронна мережа, глибокі нейронні мережі, нейроні мережі, нейронні мережи, нейронна мережа, нейро-фаззі мережі, активація нейронних мереж, модель нейронної мережі, нейронні мережі глибинного навчання, машинне навчання
• Main Author: Panchuk, B.O.
• Format: Article
• Language: Ukrainian
• Published: PROBLEMS IN PROGRAMMING 2024
• Subjects: neural networks; formal verification; satisfiability modulo theories; constraints satisfaction; intrusion detection systems; network traffic analysis; UDC 004.415.52; 004.89; 004.492
• Online Access: https://pp.isofts.kiev.ua/index.php/ojs1/article/view/644
• Journal Title: Problems in programming

This paper introduces a method for the formal verification of neural networks using a Satisfiability Modulo Theories (SMT) solver. This approach enables the mathematical validation of specific neural network properties, enhancing their predictability. We propose a method for simplifying a neural net...