New PLAA method enhances adversarial attacks on network intrusion detection systems

Researchers have developed a new method called PLAA to create adversarial attacks specifically for network intrusion detection systems (NIDS). Unlike previous methods that adapted attacks from computer vision, PLAA focuses on generating packet-level features to construct adversarial traffic. This approach ensures the generated traffic remains valid and retains its original malicious semantics. The PLAA method demonstrated a high evasion success rate of 92.78% on several NIDS models and datasets, while preserving the integrity of the adversarial traffic. AI

IMPACT This research highlights vulnerabilities in AI-powered NIDS and could lead to more robust security measures.

RANK_REASON The cluster contains a research paper detailing a new method for adversarial attacks on network intrusion detection systems. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• CIC-DDoS2019
• CIC-IDS-2017
• CIC-UNSW-NB15
• Computer vision
• Deep neural networks
• Network-based Intrusion Detection System
• PLAA

• paper
• safety

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →