Deep Learning on Real-World Graphs
It addresses scalability and practical challenges for applying GNNs in industrial settings like social networks and recommender systems, though it appears incremental as it builds on existing GNN frameworks.
This thesis tackled the limitations of Graph Neural Networks (GNNs) in real-world applications by introducing models like SIGN, TGN, Dir-GNN, Feature Propagation, and NuGget to address scalability, temporality, directionality, data incompleteness, and structural uncertainty, enabling their use in domains such as social and recommender systems.
Graph Neural Networks (GNNs) have become a central tool for learning on graph-structured data, yet their applicability to real-world systems remains limited by key challenges such as scalability, temporality, directionality, data incompleteness, and structural uncertainty. This thesis introduces a series of models addressing these limitations: SIGN for scalable graph learning, TGN for temporal graphs, Dir-GNN for directed and heterophilic networks, Feature Propagation (FP) for learning with missing node features, and NuGget for game-theoretic structural inference. Together, these contributions bridge the gap between academic benchmarks and industrial-scale graphs, enabling the use of GNNs in domains such as social and recommender systems.