Dr. Konstantinos Kyriakopoulos is a Senior Lecturer in Cyber Intelligence, specializing in cybersecurity challenges and AI-driven solutions for converged Information and Operational Technology infrastructures. His work integrates machine learning with cyber-physical systems to enhance decision-making under uncertainty, with a focus on resilient AI algorithms and autonomous defense mechanisms. He has contributed to projects funded by Dstl, Innovate UK, and industry partners, addressing Industry 4.0 security, smart vehicles, electromagnetic attacks, and smart buildings. His research interests include adversarial machine learning, evidence theory for uncertainty modeling in ML, and autonomous cyber defense systems. He has developed novel approaches using reinforcement learning, hidden Markov models, and Bayesian optimization. His work emphasizes creating ML algorithms resistant to adversarial attacks while improving cybersecurity resilience in dynamic environments. Dr. Kyriakopoulos has also explored applications of AI in educational technology, such as automated assessment of spoken English proficiency. His recent publications focus on federated learning security, causal reinforcement learning, and anomaly detection techniques. He has experience in translating research into practical solutions, including business model pitches to investors in Canary Wharf for threat modeling tools. His contributions span over 40 peer-reviewed articles, with notable work in network intrusion detection, attack prediction via HMMs, and wavelet-based compression for network measurements. His expertise bridges academia and industry, addressing both theoretical advancements and real-world implementation challenges.
