ANAnastasiya Grigorevna Nikolskaya

Anastasiya Grigorevna Nikolskaya

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Anastasiya Grigorevna Nikolskaya has a total of 4 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • On the Formalization of Constrained Adversarial Attacks in Network Traffic Analysis

    by A. G. Nikolskaya, D. A. Rybolovlev

    This paper explores the challenges in the formalization of constrained adversarial attacks against machine learning-based intrusion detection systems for network traffic analysis. While effective at detecting anomalies and zero-day attacks in network traffic, such systems are vulnerable to adversarial attacks, particularly evasion attacks that introduce carefully designed perturbations into input data to cause misclassification. The structured and heterogeneous nature of network traffic and the complexity of extracted traffic features complicate the generation of functionally valid adversarial examples that represent realistic attack scenarios. Crafting realistic adversarial examples requires incorporating domain-specific constraints into the attack. This transforms the underlying optimization problem into a constrained optimization problem whose solution is non-trivial and does not guarantee convergence. This paper reviews existing approaches to incorporating domain constraints, discusses the limitations of traditional distance metrics in the context of network traffic, and highlights challenges arising from differences between feature-space and traffic-space attacks. We conclude that domain constraints must be accounted for as comprehensively as possible while preserving practical applicability, and that the effectiveness of constrained attack methods and constraint-handling approaches remains an open question.

  • Deep Learning Applications for Intrusion Detection in Network Traffic

    by A. I. Getman, D. A. Rybolovlev, A. G. Nikolskaya

    This paper discusses the problems of applying deep learning methods for intrusion detection in network traffic. The results of analyzing the relevant studies and reviews of deep learning applications for intrusion detection are presented. The most popular deep learning methods are discussed and compared. A classification system of deep learning methods for intrusion detection is proposed. The current trends and challenges of applying deep learning methods for intrusion detection in network traffic are identified. To assess the applicability of deep learning methods for intrusion detection, the CNN-BiLSTM neural network is synthesized. The synthesized network is compared with the previously developed model based on the random forest classifier. The deep learning method makes it possible to simplify the feature engineering stage, with the values of quality metrics for the random forest and CNN-BiLSTM models being close. This confirms the high prospects for application of deep learning methods to intrusion detection.


  • Open Access

    Deep Learning Applications for Intrusion Detection in Network Traffic

    by Aleksandr Igorevich Getman, Maxim Nikolaevich Goryunov, Andrey Georgievich Matskevich, Dmitry Aleksandrovich Rybolovlev, Anastasiya Grigorevna Nikolskaya

    В работе рассмотрены вопросы применения методов глубокого обучения для обнаружения компьютерных атак в сетевом трафике. Представлены результаты анализа релевантных исследований и обзоров в области применения глубокого обучения для обнаружения вторжений. Произведено описание и сравнение наиболее используемых методов глубокого обучения, предложена система их классификации. Определены существующие тенденции и проблемы применения методов глубокого обучения для обнаружения компьютерных атак в сетевом трафике. Для оценки применимости методов глубокого обучения для обнаружения вторжений синтезирована нейронная сеть CNN-BiLSTM и представлены результаты её сравнения с разработанной ранее моделью, основанной на использовании классификатора типа «случайный лес». Использование метода глубокого обучения позволило упростить этап конструирования признаков, что вместе с близостью полученных значений метрик для сравниваемых моделей подтверждает перспективность применения методов глубокого обучения для обнаружения вторжений.