9004 Problems of information security. Computer systems Проблемы информационной безопасности. Компьютерные системы 2071-8217 4 Adversarial attacks on intrusion detection systems using LSTM classifier Состязательные атаки на системы обнаружения вторжений, использующих LSTM-классификатор Kulikov D. 0000-0002-9899-2778 Platonov Vladimir plato@ibks.spbstu.ru Peter the Great St. Petersburg Polytechnic University 03 06 2021 2 48 56

This paper discusses adversarial attacks on machine learning models and their classification. Methods for assessing the resistance of an LSTM classifier to adversarial attacks are investigated. JSMA and FGSM attacks, chosen due to the portability of adversarial examples between machine learning models, are discussed in detail. An attack of “poisoning” of the LSTM classifier is proposed. Methods of protection against the considered adversarial attacks are formulated

 adversarial attack intrusion detection system neural network LSTM