Recent advances in multi-agent reinforcement learning (MARL) have created opportunities to solve complex real-world tasks. Cybersecurity is a notable application area, where defending networks against sophisticated adversaries remains a challenging task typically performed by teams of security operators. In this work, we explore novel MARL strategies for building autonomous cyber network defenses that address challenges such as large policy spaces, partial observability, and stealthy, deceptive adversarial strategies. We propose a hierarchical Proximal Policy Optimization (PPO) architecture that decomposes the cyber defense task into specific sub-tasks like network investigation and host recovery, with sub-policies leveraged by a master defense policy for coordination. We conduct extensive experiments using CybORG Cage 4, achieving top performance in convergence speed, episodic return, and interpretable cybersecurity metrics.

Biometric Recognition is a widely used system in security systems around the world. ECG based biometric systems are not widely in use. In this paper, we propose a new way to build such a system which can either identify or verify a person's identity. We have tested our system using ECG data provided in PhysioBank ATM.