Message passing paradigms, Graph Convolutional Networks (GCN), GraphSAGE, and non-Euclidean graph representations.
For a small example, three people form a small friendship graph. Aggregate neighbor features for the center node, transform them, and compare sum, mean, and normalized updates.
Accounts and transfers form a heterogeneous graph. Create time-safe neighborhoods, train against a baseline, evaluate cold nodes, and inspect whether labels leak through edges.
Initializing Interactive Playground...
Predict how one control changes the Graph Neural Networks (GNNs) result.
The starter deliberately follows a tempting but flawed assumption: Creating edges that leak the target
Check: Your saved hypothesis predicts a direction, the experiment tests it, and the result changes or supports your explanation.
Explanations after each answer · 80% to pass · Graph Neural Networks (GNNs)
Chapter 1 of 7 · Core
Standard convolutional neural networks (CNNs) operate on regular 2D/3D grids (images/voxels) with defined spatial ordering. Graph Neural Networks (GNNs) operate on non-Euclidean graph topologies defined by node features and an adjacency matrix .
Key equations
Two free references selected for Graph Neural Networks (GNNs).
Free Paper
The original GCN paper and a compact derivation of neighborhood aggregation.
Free Documentation
Official practical introduction to graph data, message passing, batching, and graph learning workflows.
Primary sources and free technical reading for Graph Neural Networks (GNNs).