3 papers
cs.LG2026
FDN: Interpretable Spatiotemporal Forecasting with Future Decomposition Networks
Nicholas Majeske, Ariful Azad
Spatiotemporal systems comprise a collection of spatially distributed yet interdependent entities each generating unique dynamic signals. Highly sophisticated methods have been pro…
cs.LG2025
Shapley-Value-Based Graph Sparsification for GNN Inference
Selahattin Akkas, Ariful Azad
Graph sparsification is a key technique for improving inference efficiency in Graph Neural Networks by removing edges with minimal impact on predictions. GNN explainability methods…
cs.LG2025
DistShap: Scalable GNN Explanations with Distributed Shapley Values
Selahattin Akkas, Aditya Devarakonda, Ariful Azad
With the growing adoption of graph neural networks (GNNs), explaining their predictions has become increasingly important. However, attributing predictions to specific edges or fea…