Formal verification of deep neural networks

• Опубліковано в: PROBLEMS IN PROGRAMMING
• Дата: 2024
• Випуск: 2-3
• Сторінки: 253-262
• ISSN: 1727-4907
• Автори та афіліації: B.O. Panchuk — Glushkov Institute of Cybernetics NAS of Ukraine
• Автор: Panchuk, B.O.
• Формат: Стаття
• Мова: Українська
• Опубліковано: PROBLEMS IN PROGRAMMING 2024
• Теми: neural networks; formal verification; satisfiability modulo theories; constraints satisfaction; intrusion detection systems; network traffic analysis; UDC 004.415.52; 004.89; 004.492
• Онлайн доступ: https://pp.isofts.kiev.ua/index.php/ojs1/article/view/644
• Назва журналу: 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...