document

Unrolling of Simplicial ElasticNet for Edge Flow Signal Reconstruction

Liu, Chengen (author), Leus, G.J.T. (author), Isufi, E. (author)

The edge flow reconstruction task consists of retreiving edge flow signals from corrupted or incomplete measurements. This is typically solved by a regularized optimization problem on higher-order networks such as simplicial complexes and the corresponding regularizers are chosen based on prior knowledge. Tailoring this prior to the setting...

journal article 2023

document

Graph-Time Convolutional Neural Networks: Architecture and Theoretical Analysis

Sabbaqi, M. (author), Isufi, E. (author)

Devising and analysing learning models for spatiotemporal network data is of importance for tasks including forecasting, anomaly detection, and multi-agent coordination, among others. Graph Convolutional Neural Networks (GCNNs) are an established approach to learn from time-invariant network data. The graph convolution operation offers a...

journal article 2023

document

Graph-Time Trend Filtering and Unrolling Network

Sabbaqi, M. (author), Isufi, E. (author)

Reconstructing missing values and removing noise from network-based multivariate time series requires developing graph-time regularizers capable of capturing their spatiotemporal behavior. However, current approaches based on joint spatiotemporal smoothness, diffusion, or variations thereof may not be effective for time series with...

conference paper 2023

document

EdgeNets: Edge Varying Graph Neural Networks

Isufi, E. (author), Gama, Fernando (author), Ribeiro, Alejandro (author)

Driven by the outstanding performance of neural networks in the structured euclidean domain, recent years have seen a surge of interest in developing neural networks for graphs and data supported on graphs. The graph is leveraged at each layer of the neural network as a parameterization to capture detail at the node level with a reduced...

journal article 2022

document

Learning Time-Varying Graphs from Online Data

Natali, A. (author), Isufi, E. (author), Coutino, Mario (author), Leus, G.J.T. (author)

This work proposes an algorithmic framework to learn time-varying graphs from online data. The generality offered by the framework renders it model-independent, i.e., it can be theoretically analyzed in its abstract formulation and then instantiated under a variety of model-dependent graph learning problems. This is possible by phrasing (time...

journal article 2022

document

Convolutional Filtering in Simplicial Complexes

Isufi, E. (author), Yang, Maosheng (author)

This paper proposes convolutional filtering for data whose structure can be modeled by a simplicial complex (SC). SCs are mathematical tools that not only capture pairwise relationships as graphs but account also for higher-order network structures. These filters are built by following the shift-and-sum principle of the convolution operation and...

conference paper 2022

document

Online Edge Flow Imputation on Networks

Money, Rohan (author), Krishnan, Joshin (author), Beferull-Lozano, Baltasar (author), Isufi, E. (author)

An online algorithm for missing data imputation for networks with signals defined on the edges is presented. Leveraging the prior knowledge intrinsic to real-world networks, we propose a bi-level optimization scheme that exploits the causal dependencies and the flow conservation, respectively via <italic>(i)</italic> a sparse line...

journal article 2022

document

Task-Aware Connectivity Learning for Incoming Nodes Over Growing Graphs

Das, B. (author), Hanjalic, A. (author), Isufi, E. (author)

Data processing over graphs is usually done on graphs of fixed size. However, graphs often grow with new nodes arriving over time. Knowing the connectivity information of these nodes, and thus, the expanded graph is crucial for processing data over the expanded graph. In its absence, its inference and the subsequent data processing become...

journal article 2022

document

Accuracy-diversity trade-off in recommender systems via graph convolutions

Isufi, E. (author), Pocchiari, Matteo (author), Hanjalic, A. (author)

Graph convolutions, in both their linear and neural network forms, have reached state-of-the-art accuracy on recommender system (RecSys) benchmarks. However, recommendation accuracy is tied with diversity in a delicate trade-off and the potential of graph convolutions to improve the latter is unexplored. Here, we develop a model that learns...

journal article 2021

document

Graph-time convolutional neural networks

Isufi, E. (author), Mazzola, Gabriele (author)

Spatiotemporal data can be represented as a process over a graph, which captures their spatial relationships either explicitly or implicitly. How to leverage such a structure for learning representations is one of the key challenges when working with graphs. In this paper, we represent the spatiotemporal relationships through product graphs...

conference paper 2021

document

Nonlinear State-Space Generalizations of Graph Convolutional Neural Networks

Ruiz, Luana (author), Gama, Fernando (author), Ribeiro, Alejandro (author), Isufi, E. (author)

Graph convolutional neural networks (GCNNs) learn compositional representations from network data by nesting linear graph convolutions into nonlinearities. In this work, we approach GCNNs from a state-space perspective revealing that the graph convolutional module is a minimalistic linear state-space model, in which the state update matrix is...

conference paper 2021

document

Topological Volterra Filters

Leus, G.J.T. (author), Yang, Maosheng (author), Coutino, Mario (author), Isufi, E. (author)

To deal with high-dimensional data, graph filters have shown their power in both graph signal processing and data science. However, graph filters process signals exploiting only pairwise interactions between the nodes, and they are not able to exploit more complicated topological structures. Graph Volterra models, on the other hand, are also...

conference paper 2021

document

Sampling Graph Signals with Sparse Dictionary Representation

Zhang, Kaiwen (author), Coutino, Mario (author), Isufi, E. (author)

Graph sampling strategies require the signal to be relatively sparse in an alternative domain, e.g. bandlimitedness for reconstructing the signal. When such a condition is violated or its approximation demands a large bandwidth, the reconstruction often comes with unsatisfactory results even with large samples. In this paper, we propose an...

conference paper 2021

document

Online Graph Learning From Time-Varying Structural Equation Models

Natali, A. (author), Isufi, E. (author), Coutino, Mario (author), Leus, G.J.T. (author)

Topology identification is an important problem across many disciplines, since it reveals pairwise interactions among entities and can be used to interpret graph data. In many scenarios, however, this (unknown) topology is time-varying, rendering the problem even harder. In this paper, we focus on a time-varying version of the structural...

conference paper 2021

document

Finite Impulse Response Filters for Simplicial Complexes

Yang, Maosheng (author), Isufi, E. (author), Schaub, Michael T. (author), Leus, G.J.T. (author)

In this paper, we study linear filters to process signals defined on simplicial complexes, i.e., signals defined on nodes, edges, triangles, etc. of a simplicial complex, thereby generalizing filtering operations for graph signals. We propose a finite impulse response filter based on the Hodge Laplacian, and demonstrate how this filter can be...

conference paper 2021

document

Towards Finite-Time Consensus with Graph Convolutional Neural Networks

Iancu, A. (author), Isufi, E. (author)

Atrial electrograms are often used to gain understanding on the development of atrial fibrillation (AF). Using such electrograms, cardiologists can reconstruct how the depolarization wave-front propagates across the atrium. Knowing the exact moment at which the depolarization wavefront in the tissue reaches each electrode is an important aspect...

conference paper 2020

document

State-space based network topology identification

Coutino, Mario (author), Isufi, E. (author), Maehara, T. (author), Leus, G.J.T. (author)

In this work, we explore the state-space formulation of network processes to recover the underlying network structure (local connections). To do so, we employ subspace techniques borrowed from system identification literature and extend them to the network topology inference problem. This approach provides a unified view of the traditional...

conference paper 2020

document

Graph-time spectral analysis for atrial fibrillation

Sun, M. (author), Isufi, E. (author), de Groot, N.M.S. (author), Hendriks, R.C. (author)

Atrial fibrillation is a clinical arrhythmia with multifactorial mechanisms still unresolved. Time-frequency analysis of epicardial electrograms has been investigated to study atrial fibrillation. However, deeper understanding can be achieved by incorporating the spatial dimension. Unfortunately, the physical models describing the spatial...

journal article 2020

document

Rational Chebyshev Graph Filters

Rimleanscaia, Oxana (author), Isufi, E. (author)

This paper proposes rational Chebyshev graph filters to approximate step graph spectral responses with arbitrary precision, which are of interest in graph filter banks and spectral clustering. The proposed method relies on the well-known Chebyshev filters of the first kind and on a domain transform of the angular frequencies to the graph...

conference paper 2020

document

State-Space Network Topology Identification from Partial Observations

Coutino, Mario (author), Isufi, E. (author), Maehara, Takanori (author), Leus, G.J.T. (author)

In this article, we explore the state-space formulation of a network process to recover from partial observations the network topology that drives its dynamics. To do so, we employ subspace techniques borrowed from system identification literature and extend them to the network topology identification problem. This approach provides a unified...

journal article 2020