Al for Networked Infrastruct. Systems: Graph Learning Resilience-Oriented Decision-Making Algorithm

Discover how cutting-edge AI is transforming the future of infrastructure in this FTS seminar featuring Dr. Xudong Fan (SUNY at Buffalo). In this talk, Dr. Fan explores how graph learning and deep reinforcement learning can optimize decision-making for complex, networked systems like roads, water distribution, and power networks. The presentation highlights real-world challenges such as aging infrastructure, system failures, and disaster recovery—and demonstrates how AI can improve resilience through smarter repair strategies, faster response times, and data-driven insights. Dr. Fan also introduces innovative approaches using graph neural networks and multimodal learning to better model spatial systems and enhance prediction accuracy. The session concludes with a forward-looking discussion on the role of generative AI, multi-agent systems, and human-centered decision-making in infrastructure management. Perfect for researchers, engineers, and anyone interested in AI applications for smart cities and resilient infrastructure systems.