Skip to content

Representation Learning

Unsupervised and self-supervised graph representation learning.

All categories

  • Inductive Deep Graph Infomax with GraphSAGE and neighbor sampling

    Learn node embeddings without labels on a graph too large to process at once.

    Cora (Planetoid) · SAGEConv · DeepGraphInfomax

    Read Tutorial    

  • Node embeddings from random walks (Node2Vec)

    Learn an embedding for every paper of the Cora graph from random walks, without looking at features or labels.

    Cora (Planetoid) · Node2Vec

    Read Tutorial    

  • Pre-trained positional and structural encodings (GPSE)

    GPSE (Cantürk et al., 2023) is a GNN pre-trained to predict many positional and structural encodings of nodes (random-walk statistics, eigenvectors, ...).

    ESOL (MoleculeNet) · GCNConv · GPSE

    Read Tutorial    

  • Unsupervised GraphSAGE (Cora)

    Learn node embeddings of the Cora graph without labels: GraphSAGE (Hamilton et al., 2017) is trained to predict which pairs of nodes are linked.

    Cora (Planetoid) · GraphSAGE

    Read Tutorial    

  • Unsupervised GraphSAGE on protein graphs (PPI)

    Learn embeddings of the proteins in the PPI graphs without labels: GraphSAGE (Hamilton et al., 2017) is trained to predict which proteins interact.

    PPI · GraphSAGE

    Read Tutorial    

  • Unsupervised node embeddings with Deep Graph Infomax (Cora)

    Learn node embeddings of the Cora citation graph without using any labels.

    Cora (Planetoid) · GCNConv · DeepGraphInfomax

    Read Tutorial