Explore a 15-minute IEEE conference talk examining multi-instance adversarial attacks on Graph Neural Network (GNN) based malicious domain detection systems. Delve into the vulnerabilities of GNN models in cybersecurity applications and learn about innovative attack strategies that exploit these weaknesses. Gain insights into the challenges faced by current malicious domain detection methods and understand the potential implications for network security. Discover cutting-edge research findings and potential countermeasures to enhance the robustness of GNN-based detection systems against sophisticated adversarial threats.
Multi-Instance Adversarial Attack on GNN-Based Malicious Domain Detection
Overview
Syllabus
211 Multi Instance Adversarial Attack on GNN Based Malicious Domain Detection Mahmoud Khaled Ahm
Taught by
IEEE Symposium on Security and Privacy
Tags
Related Courses
-
Multi-Instance Adversarial Attack on GNN-Based Malicious Domain Detection
-
Penetration Testing and Ethical Hacking
-
Exploring Adversarial Examples in Malware Detection
-
Trust Region Based Adversarial Attack on Neural Networks
-
Perception Deception - Physical Adversarial Attack Challenges
-
Untargeted Adversarial Attack on Knowledge Graph Embeddings
