CD
C. De Bacco
info
Please Note
<p>This page displays the records of the person named above and is not linked to a unique person identifier. This record may need to be merged to a profile.</p>
2 records found
1
Graphs are a powerful data structure for representing relational data and are widely used to describe complex real-world systems. Probabilistic graphical models (PGMs) and graph neural networks (GNNs) can both leverage graph-structured data, but their inherent functioning is different. The question is how do they compare in capturing the information contained in networked datasets? We address this objective by solving a link prediction task and we conduct three main experiments, on both synthetic and real networks. The first focuses on how PGMs and GNNs handle input features, considering both their type and dimensionality. The second investigates their robustness to noise injected either into node features or graph edges, while the third investigates the impact of graph heterophily. PGMs do not necessarily require features on nodes, while GNNs cannot exploit the network edges alone, and the choice of input features matters. We compare standard GNNs architectures with three PGMs models, variants of stochastic block models. We find that GNNs are outperformed by PGMs when input features are low-dimensional or noisy, mimicking many real scenarios where node attributes might be scalar, missing or corrupted. When high-quality, high-dimensional input features are available and the graph is assortative, both methods can achieve comparable performance on average, but with GNNs outperforming PGMs on a fold-by-fold basis on real networks. Then, we find that PGMs are more robust than GNNs when the heterophily of the graph is increased. Introducing noise into the edges amplifies these trends: the GNN advantage on real assortative graphs diminishes, while the regimes in which PGMs outperform GNNs become even more pronounced. Finally, to assess performance beyond prediction tasks, we also compare the two frameworks in terms of their computational complexity and interpretability.
...
Graphs are a powerful data structure for representing relational data and are widely used to describe complex real-world systems. Probabilistic graphical models (PGMs) and graph neural networks (GNNs) can both leverage graph-structured data, but their inherent functioning is different. The question is how do they compare in capturing the information contained in networked datasets? We address this objective by solving a link prediction task and we conduct three main experiments, on both synthetic and real networks. The first focuses on how PGMs and GNNs handle input features, considering both their type and dimensionality. The second investigates their robustness to noise injected either into node features or graph edges, while the third investigates the impact of graph heterophily. PGMs do not necessarily require features on nodes, while GNNs cannot exploit the network edges alone, and the choice of input features matters. We compare standard GNNs architectures with three PGMs models, variants of stochastic block models. We find that GNNs are outperformed by PGMs when input features are low-dimensional or noisy, mimicking many real scenarios where node attributes might be scalar, missing or corrupted. When high-quality, high-dimensional input features are available and the graph is assortative, both methods can achieve comparable performance on average, but with GNNs outperforming PGMs on a fold-by-fold basis on real networks. Then, we find that PGMs are more robust than GNNs when the heterophily of the graph is increased. Introducing noise into the edges amplifies these trends: the GNN advantage on real assortative graphs diminishes, while the regimes in which PGMs outperform GNNs become even more pronounced. Finally, to assess performance beyond prediction tasks, we also compare the two frameworks in terms of their computational complexity and interpretability.
Complex systems are often driven by higher-order interactions among multiple units, naturally represented as hypergraphs. Understanding dependency structures within these hypergraphs is crucial for understanding and predicting the behavior of complex systems but is made challenging by their combinatorial complexity and computational demands. In this paper, we introduce a class of probabilistic models that efficiently represents and discovers a broad spectrum of mesoscale structure in large-scale hypergraphs. The key insight enabling this approach is to treat classes of similar units as themselves nodes in a latent hypergraph. By modeling observed node interactions through latent interactions among classes using low-rank representations, our approach tractably captures rich structural patterns while ensuring model identifiability. This allows for direct interpretation of distinct node- and class-level structures. Empirically, our model improves link prediction over state-of-the-art methods and discovers interpretable structures in diverse real-world systems, including pharmacological and social networks, advancing our ability to incorporate large-scale higher-order data into the scientific process.
...
Complex systems are often driven by higher-order interactions among multiple units, naturally represented as hypergraphs. Understanding dependency structures within these hypergraphs is crucial for understanding and predicting the behavior of complex systems but is made challenging by their combinatorial complexity and computational demands. In this paper, we introduce a class of probabilistic models that efficiently represents and discovers a broad spectrum of mesoscale structure in large-scale hypergraphs. The key insight enabling this approach is to treat classes of similar units as themselves nodes in a latent hypergraph. By modeling observed node interactions through latent interactions among classes using low-rank representations, our approach tractably captures rich structural patterns while ensuring model identifiability. This allows for direct interpretation of distinct node- and class-level structures. Empirically, our model improves link prediction over state-of-the-art methods and discovers interpretable structures in diverse real-world systems, including pharmacological and social networks, advancing our ability to incorporate large-scale higher-order data into the scientific process.