Editing Graph Neural Networks (section)

== <span style="color: #FFFFFF;">Remembering</span> ==
* '''Graph''' — A mathematical structure consisting of nodes (vertices) connected by edges. Edges can be directed or undirected, weighted or unweighted.
* '''Node''' — An entity in a graph (a person in a social network, an atom in a molecule, a paper in a citation network).
* '''Edge''' — A relationship between two nodes (friendship, chemical bond, citation, road connection).
* '''Node features''' — Attribute vectors associated with each node (e.g., user profile features, atom type, paper topic).
* '''Edge features''' — Attribute vectors associated with each edge (e.g., bond type, relationship strength, road distance).
* '''Adjacency matrix''' — A square matrix A where A[i][j] = 1 if there is an edge from node i to node j.
* '''Neighborhood''' — The set of nodes directly connected to a given node by edges.
* '''Message passing''' — The core GNN operation: each node sends "messages" (feature vectors) to its neighbors, which aggregate them to update node representations.
* '''Node-level task''' — Predicting a property of each node (e.g., classifying users as spammers or not).
* '''Edge-level task''' — Predicting properties of or between pairs of nodes (e.g., link prediction: will user A befriend user B?).
* '''Graph-level task''' — Predicting a property of the entire graph (e.g., classifying a molecule's toxicity).
* '''GCN (Graph Convolutional Network)''' — A foundational GNN that aggregates node features from neighbors with normalized averaging.
* '''GraphSAGE''' — A GNN that samples a fixed number of neighbors for scalable inductive learning on large graphs.
* '''GAT (Graph Attention Network)''' — A GNN that learns attention weights for neighbor aggregation, giving more weight to important neighbors.
* '''Readout/Pooling''' — Aggregating all node representations into a single graph-level representation for graph classification.
</div>

<div style="background-color: #006400; color: #FFFFFF; padding: 20px; border-radius: 8px; margin-bottom: 15px;">