6 papers
Hyperbolic Graph Representation Learning: Embed in One Metric, Optimize with Another
Federico Larroca, Paola Bermolen, Marcelo Fiori +1
Hierarchical graphs embed in hyperbolic space with lower distortion than in Euclidean space owing to its negative curvature. However, their gradient-based learning is hampered at l…
A Unified Framework of Hyperbolic Graph Representation Learning Methods
Sofía Pérez Casulo, Marcelo Fiori, Bernardo Marenco +1
Hyperbolic geometry has emerged as an effective latent space for representing complex networks, owing to its ability to capture hierarchical organization and heterogeneous connecti…
GNNs for Time Series Anomaly Detection: An Open-Source Framework and a Critical Evaluation
Federico Bello, Gonzalo Chiarlone, Marcelo Fiori +2
There is growing interest in applying graph-based methods to Time Series Anomaly Detection (TSAD), particularly Graph Neural Networks (GNNs), as they naturally model dependencies a…
Weighted Random Dot Product Graphs
Bernardo Marenco, Paola Bermolen, Marcelo Fiori +2
Modeling of intricate relational patterns has become a cornerstone of contemporary statistical research and related data science fields. Networks, represented as graphs, offer a na…
LASE: Learned Adjacency Spectral Embeddings
Sofía Pérez Casulo, Marcelo Fiori, Federico Larroca +1
We put forth a principled design of a neural architecture to learn nodal Adjacency Spectral Embeddings (ASE) from graph inputs. By bringing to bear the gradient descent (GD) method…
Some families of digraphs determined by the complementarity spectrum
Diego Bravo, Florencia Cubría, Marcelo Fiori +1
We examine the capacity of the complementarity spectrum to distinguish non-isomorphic digraphs. We focus on the seven families with exactly three complementarity eigenvalues. Our f…