As the Internet of Things (IoT) becoming the mordern technology trend and is applied in self-driving automobiles, smart healthcare, and smart cities, IoT vulnerabilities and the intrusion threats it faces can lead to considerable damage to devices or the system. IoT intrusion detection systems (IDSs) must be researched and enhanced to adapt to modern network attacks. This paper proposes a Multi-task Deep Learning-based approach for network traffic classification for IoT environment for detection of rare network intrusions with sparse data, which can outperform Single-task Deep Learning models for intrusion detection in IoT network.
Дун Х. (науч. рук. Котенко И.В.) MULTI-TASK DEEP LEARNING-BASED INTRUSION DETECTION MODEL FOR IOT NETWORK // Сборник тезисов докладов конгресса молодых ученых. Электронное издание. – СПб: Университет ИТМО, [2023]. URL: https://kmu.itmo.ru/digests/article/9988