Graphs & Relational Learning

We explore graph-based machine learning methods for modeling relationships and interactions.

Topics

Scalable GNN architectures, expressive message passing, and large-scale graph learning.

Friend recommendation, connection prediction, and structural bias in graph models.

Detecting fraud, spam, and unusual behavior in social networks and interaction graphs.

Zero-shot and few-shot reasoning over text-attributed graphs using large language models.

Data quality-aware graph learning and fairness in graph-based predictions.

Self-supervised, contrastive, and scalable methods for learning on large graphs.