Collaborative Research: Network Control Systems Science for Graph Machine Learning

Understanding complex networks—such as social networks, transportation systems, or communication infrastructures—is essential for many modern technologies. This project explores a new way to analyze these networks by combining ideas from control systems and machine learning to better understand how they behave and how they can be used for prediction and decision-making.

Instead of looking at networks as static structures, the research treats them as dynamic systems that can be “tested” by introducing signals and observing how they respond. This approach provides deeper insight into how different parts of a network are connected and interact. By capturing these patterns more effectively, the project aims to improve how machines learn from network data—supporting tasks like identifying important connections, classifying network structures, and predicting how networks will evolve.

Ultimately, this work offers a new perspective on analyzing complex systems, with potential applications across fields such as cybersecurity, infrastructure management, and data science.