and illuminate the most probable adversarial attack paths in your networks. Containing entirely original research, this book outlines findings and real-world scenarios that have been modeled and tested against custom generated networks, simulated networks, and data. You'll also find: * A thorough introduction to modeling actions within post-exploitation cybersecurity events, including Markov Decision Processes employing warm-up phases and penalty scaling * Comprehensive explorations of penetration testing automation, including how RL is trained and tested over a standard attack graph construct * Practical discussions of both red and blue team objectives in their efforts to exploit and defend networks, respectively * Complete treatment of how reinforcement learning can be applied to real-world cybersecurity operational scenarios Perfect for practitioners working in cybersecurity, including cyber defenders and planners, network administrators, and information security professionals, Reinforcement Learning for Cyber Operations: Applications of Artificial Intelligence for Penetration Testing will also benefit computer science researchers. |