Towards a Graph Neural Network-Based Approach for Estimating Hidden States in Cyber Attack Simulations

2023/12/09 by Pontus Johnson, Johnson, Pontus, Mathias Ekstedt +1
Computer Science · Engineering · #Cryptography and Security (cs.CR) #FOS: Computer and information sciences #Information and Cyber Security #Network Security and Intrusion Detection #Smart Grid Security and Resilience

paper · pdf · doi:10.48550/arxiv.2312.05666

openalex publication_date 2023/12/09 · openalex created_date 2023/12/13 · openalex updated_date 2026/07/28

Abstract

This work-in-progress paper introduces a prototype for a novel Graph Neural Network (GNN) based approach to estimate hidden states in cyber attack simulations. Utilizing the Meta Attack Language (MAL) in conjunction with Relational Dynamic Decision Language (RDDL) conformant simulations, our framework aims to map the intricate complexity of cyber attacks with a vast number of possible vectors in the simulations. While the prototype is yet to be completed and validated, we discuss its foundational concepts, the architecture, and the potential implications for the field of computer security.

Related

• Use of Graph Neural Networks in Aiding Defensive Cyber Operations
• DeepHunter: A Graph Neural Network Based Approach for Robust Cyber Threat Hunting
• Illuminati: Towards Explaining Graph Neural Networks for Cybersecurity Analysis
• An Automated, End-to-End Framework for Modeling Attacks From Vulnerability Descriptions
• AI-based Attack Graph Generation