Predicting steady-state behavior in complex networks with graph neural networks.

In complex systems, information propagation can be defined as diffused or delocalized, weakly localized, and strongly localized. This study investigates the application of graph neural network models to learn the behavior of a linear dynamical system…

Preview only. Read the full abstract at the source