Big data network

Networks are ubiquitous in our modern vocabulary. From social media to transportation and resource distribution networks, they are the fabric of the interdependent infrastructures we rely on to navigate the complexities of our daily lives. Our work streamlines network analysis, offering effective tools to assist policymakers as they grapple with the most pressing issues: How does a city recover from a natural disaster? How does climate change impact society? How can we best fight infectious disease outbreaks? The versatile tools we have developed interpret, forecast, and explain the dynamics of massively interacting systems.

 

 

Graph Dynamical Systems

Graph Dynamical Systems (GDS) provide a mathematically rigorous framework for modeling how local interactions on networks generate large-scale system behavior. By coupling network structure with formally defined node-level update rules, GDS enables us to analyze, forecast, and control dynamics in biological, social, and infrastructure systems.

Our work spans foundational theory, algorithm design, learning and inference, and scalable high-performance simulation, forming a core pillar of our Network Science program.

 

Selected Foundational and High-Impact Publications in Graph Dynamical Systems

Foundational Theory of Graph Dynamical Systems

  • Barrett C, Hunt III HB, Marathe M, Ravi S, Rosenkrantz D, Stearns R (2003) On some special classes of sequential dynamical systems. Annals of Combinatorics, 7(4):381-408.
  • Mortveit H, Reidys C (2008) An Introduction to Sequential Dynamical Systems. Springer Verlag.  

Mathematical Structure & Phase Space Results

  • Kuhlman C, Mortveit H (2014) Attractor Stability in Nonuniform Boolean Networks. Theoretical Computer Science, 559:20-33.
  • Eubank S, Youssef M, Khorramzadeh Y (2014) Using the network reliability polynomial to characterize and design networks. Journal of Complex Networks, 2(4):356-372.

Learning and Inference (AI + Theory)

  • Adiga A, Kuhlman C, Marathe M, Ravi SS, Vullikanti A (2019) PAC learnability of node functions in networked dynamical systems. Proceedings of the International Conference on Machine Learning (ICML-2019):82-91.
  • Qiu Z, Adiga A, Marathe M, Ravi SS, Rosenkrantz DJ, Stearns RE, Vullikanti A (2024) Efficient PAC Learnability of Dynamical Systems Over Multilayer Networks. Proceedings of the International Conference on Machine Learning (ICML-2024):1–25.

Fundamental Limits & Theoretical Boundaries

  • Rosenkrantz DJ, Vullikanti A, Ravi SS, Stearns RE, Levin S, Poor HV, Marathe M (2022). Fundamental limitations on efficiently forecasting certain epidemic measures in network models. Proceedings of the National Academy of Sciences, 119(4):p.e2109228119.

High-Impact Applications Powered by GDS

  • Eubank S, Guclu H, Kumar A, Marathe M, Srinivasan A, Toroczkai Z, Wang N (2004) Modelling disease outbreaks in realistic urban social networks. Nature, 429(6988):180-184.
  • Chen J, Hoops S, Mortveit H, Lewis B, Machi D, Bhattacharya P, Venkatramanan S, Wilson M, Barrett C, Marathe M (2025) Epihiper—A high performance computational modeling framework to support epidemic science. PNAS nexus, 4(1):557.

Projects

Image
AIIA logo
Active
Image
Networks across the city
Active
Image
fir tree
Completed