FCLG combines local graph augmentation contrastive learning with model-level contrastive distillation to learn unsupervised graph embeddings in a federated setting, beating InfoGraph and MVGRL baselines on graph clustering.
Moleculenet: A benchmark for molecular machine learning,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Federated Contrastive Learning of Graph-Level Representations
FCLG combines local graph augmentation contrastive learning with model-level contrastive distillation to learn unsupervised graph embeddings in a federated setting, beating InfoGraph and MVGRL baselines on graph clustering.