Interpreting Temporal Graph Neural Networks with Koopman Theory
This paper presents a novel approach to interpret temporal graph models using Koopman theory.
Research projects
Probabilistic load forecasting with Reservoir Computing
This work rigorously explores the compatibility between some of the most used methods of Uncertainty Quantification with reservoir computing.
Explainability in subgraphs-enhanced Graph Neural Networks
Here we extended the use of PGExplainer, a popular XAI method for GNNs, to subgraphs-enhanced GNNs, a particular architecture designed to increase the expres...
Combining Stochastic Explainers and Subgraph Neural Networks can Increase Expressivity and Interpretability
We designed a subgraph-enhanced GNN, which learns to sample meaningful subgraphs that work both as explanations and to increase the expressive power of the m...