GINA: Neural Relational Inference From Independent Snapshots
This addresses the challenge of reconstructing interaction networks in complex systems where dynamics are unknown and data may be independent, with potential applications in fields like biology and social sciences, though it appears incremental as it builds on existing graph inference methods.
The paper tackles the problem of inferring unknown interaction structures in dynamical systems from independent snapshots of agent states, proposing GINA, a graph neural network that simultaneously learns the latent interaction graph and predicts node states, demonstrating effectiveness across various graphs and processes.
Dynamical systems in which local interactions among agents give rise to complex emerging phenomena are ubiquitous in nature and society. This work explores the problem of inferring the unknown interaction structure (represented as a graph) of such a system from measurements of its constituent agents or individual components (represented as nodes). We consider a setting where the underlying dynamical model is unknown and where different measurements (i.e., snapshots) may be independent (e.g., may stem from different experiments). We propose GINA (Graph Inference Network Architecture), a graph neural network (GNN) to simultaneously learn the latent interaction graph and, conditioned on the interaction graph, the prediction of a node's observable state based on adjacent vertices. GINA is based on the hypothesis that the ground truth interaction graph -- among all other potential graphs -- allows to predict the state of a node, given the states of its neighbors, with the highest accuracy. We test this hypothesis and demonstrate GINA's effectiveness on a wide range of interaction graphs and dynamical processes.