Complex systems consist of interacting components that produce emergent behaviors difficult to predict through traditional methods. While statistical physics and network science have provided valuable insights, they often fail to capture the intricate dependencies of real-world networks.

Recent AI advancements, particularly Graph Neural Networks (GNNs) and graph-based deep learning, offer powerful data-driven approaches for modeling complex systems. These techniques provide flexible, scalable representations that adapt to evolving interactions and preserve topological properties, enhancing our ability to predict dynamics and uncover latent structures.

This workshop aims to advance research on applying AI to complex systems interpretation across diverse domains including biological networks, social systems, urban infrastructure, and financial markets. We focus on recent advances in graph structure learning, network representation, and probabilistic generative models.

We invite contributions that:

• Advance theoretical understanding of complex systems through AI

• Develop novel computational methods for network analysis

• Demonstrate innovative real-world applications

The event will feature discussions on current challenges including model interpretability, scalability of graph-based approaches, and integration of domain knowledge into machine learning frameworks.

This satellite event brings together experts from complex systems, network science, and machine learning to exchange ideas, establish collaborations, and address open challenges in complex systems using AI techniques.