Skip to content
AI360XpertBeta
Glossary
Definition

Graph Neural Networks

A network that builds each node's representation by repeatedly combining it with messages gathered from its graph neighbors, learning from topological relationships rather than isolated rows.

Think of It Like This

Like gossip spreading through an office where each coworker updates their understanding of an announcement after chatting with their desk neighbors.

Each layer gathers messages from a node's neighbors and combines them with the node's own state to produce its next representation, and stacking layers lets information travel further across the graph. Past roughly three to four layers, this repeated averaging tends to make every node's representation converge toward the same value, a failure called over-smoothing.

GCN, GraphSAGE, and GAT differ mainly in how they aggregate neighbor messages — fixed normalized averaging, sampled subsets for scale, or learned attention weights, respectively.