arXiv subjects

Elvin Isufi

Publications and source records attributed to Elvin Isufi.

At least 19 recordsLinked to original sources

arXiv2026

Learning with Covariance Matrices: Principal Component Analysis Meets Learning with Graphs

This feature article provides an overview of the theoretical foundations for coVariance neural networks (VNNs), i.e., graph neural networks (GNNs) operating on covariance matrices as graphs. Covariance matrices are ubiquitous across domains, and hence, the deployment of GNNs often leverages graphs of pairwise statistical dependencies. Existing theoretical contributions on GNNs consider abstract graph representations and cannot accommodate the data-driven nuances associated with covariance matrices. This tutorial brings into focus various novel theoretical insights via mathematical analyses of VNNs that have broad signal processing implications, including: (i) a conceptual equivalence between VNNs and principal component analysis (PCA)-based information processing; (ii) refined stability bounds on predictive outcomes in the presence of finite sample-induced covariance matrix perturbations; and (iii) refined characterization of transferability of VNNs across multiscale datasets. The theoretical insights discussed herein provide the underlying principles and justification towards adopting VNNs over workhorse PCA-based learning pipelines, in applications where covariance matrices are useful descriptors of data structure. We also convey how impact of these foundational advances permeates to \textit{principled} designs and applications of learning methods across broad domains where covariance matrices emerge. Notably, we elucidate the conceptual insights facilitated by VNNs to the specific task of characterizing brain age gap for neurodegenerative conditions using neuroimaging datasets, a timely problem in computational neuroscience. Broader impacts to other application domains are discussed as well.

cs.LG↗

arXiv2026

Learning the Topology of a Simplicial Complex Using Noisy Simplicial Signals

Graphs are a fundamental tool for modeling the irregular (non-Euclidean) structure of complex data. However, they are inherently limited to representing pairwise relationships, making them inadequate for datasets exhibiting higher-order interactions. Simplicial complexes (SCs) have emerged as a promising framework for capturing such higher-order dependencies. This paper focuses on the problem of identifying the topology of an SC from signals, which serves as the foundation for SC-based processing and learning schemes. We consider a setting where we observe noisy signals (features) associated with the nodes of the SC (0-simplices) and a subset of the edges (1-simplices). We assume the observed signals are smooth over the unknown SC topology, and that the higher-order interactions are sparse. Building on these assumptions, we formulate topology learning as a nonconvex optimization problem and propose an efficient block-coordinate descent (BCD) algorithm to solve it. A key step in our formulation is the modeling of the topology of the SC using binary edge and triangle selection vectors, combined with efficient greedy algorithms for optimizing such vectors. We establish theoretical convergence guarantees to a stationary point of a relaxed (penalized) version of the problem and discuss computational complexity. Multiple numerical experiments with both synthetic and real-world datasets validate the effectiveness of our approach, highlighting the capability of SC-learning methods to uncover and model higher-order relationships in complex datasets.

eess.SP↗

arXiv2026

Exact Bayesian Tracking of Dynamic Network Topologies

Tracking the temporal evolution of network topologies is a fundamental challenge in social networks, epidemiology, and sensor systems, among others. This paper develops an exact Bayesian tracker for unweighted, directed graphs using nodal signal observations. This framework yields the full posterior probability distribution over network states at each time step, naturally enabling uncertainty quantification, prediction, and principled decision-making. We model the network dynamics as a Markov process on the Boolean hypercube, where edges transition independently according to a flip probability. For efficient computation, we cast the prediction step as a dyadic convolution, and leverage the Fast Walsh-Hadamard Transform to reduce the computational cost from $\mathcal{O} (4^k)$ to $\mathcal{O} (k 2^k)$, where $k$ is the maximum node degree. When the network transition probabilities are unknown, we develop an Expectation-Maximization framework to learn them from the observed signals. Comprehensive experiments on synthetic and six real-world datasets validate the proposed method and demonstrate its superior tracking accuracy, faster recovery from topological changes, and meaningful uncertainty estimates compared to state-of-the-art and classical baselines.

eess.SP↗

arXiv2026

Sparse Covariance Neural Networks

Covariance Neural Networks (VNNs) perform graph convolutions on the covariance matrix of input data to leverage correlation information as pairwise connections. They have achieved success in a multitude of applications such as neuroscience, financial forecasting, and sensor networks. However, the empirical covariance matrix on which VNNs operate typically contains spurious correlations, creating a mismatch with the actual covariance matrix that degrades VNNs' performance and computational efficiency. To tackle this issue, we put forth Sparse coVariance Neural Networks (S-VNNs), a framework that applies sparsification techniques on the sample covariance matrix and incorporates the latter into the VNN architecture. We investigate the S-VNN when the underlying data covariance matrix is both sparse and dense. When the true covariance matrix is sparse, we propose hard and soft thresholding to improve the covariance estimation and reduce the computational cost. Instead, when the true covariance is dense, we propose a stochastic sparsification where data correlations are dropped in probability according to principled strategies. Besides performance and computation improvements, we show that S-VNNs are more stable to finite-sample covariance estimations than nominal VNNs and the analogous sparse principal component analysis. By analyzing the impact of sparsification on their behavior, we tie the S-VNN stability to the data distribution and sparsification approach. We support our theoretical findings with experimental results on a variety of application scenarios, ranging from brain data to human action recognition, and show an improved task performance, improved stability, and reduced computational time compared to alternatives.

cs.LG↗

arXiv2026

Q-Net: Queue Length Estimation via Kalman-based Neural Networks

Estimating queue lengths at signalized intersections is a long-standing challenge in traffic management. Partial observability of vehicle flows complicates this task despite the availability of two privacy-preserving data sources: (i) aggregated vehicle counts from loop detectors near stop lines, and (ii) aggregated floating car data (aFCD) that provide segment-wise average speed measurements. However, how to integrate these sources with differing spatial and temporal resolutions for queue length estimation is rather unclear. Addressing this question, we present Q-Net: a queue estimation framework built upon a state-space formulation. This design addresses key challenges in queue modeling, such as violations of traffic conservation assumptions. Q-Net follows the Kalman predict-update structure and maintains physical interpretability in both the state evolution and measurement models. Q-Net uses an AI-augmented Kalman filter to learn time-varying gain dynamics from data. The framework supports real-time implementation and improves spatial transferability by grouping aFCD measurements into fixed-size local groups, making the number of learnable parameters independent of section length. Evaluations on urban main roads in Rotterdam, the Netherlands, show that Q-Net outperforms baseline methods, tracks queue formation and dissipation accurately, and mitigates aFCD-induced delays. By combining data efficiency, interpretability, real-time applicability, and spatial transferability, Q-Net makes accurate queue length estimation possible without costly sensing infrastructure like cameras or radar.

cs.LG↗

arXiv2026

Flow Matching Policy Optimization with Mirror Descent and Entropy Constraints

Balancing policy expressiveness with the exploration-exploitation trade-off is a core challenge in online Reinforcement Learning (RL). While Stochastic Differential Equation (SDE)-based diffusion policies can represent complex, multimodal action distributions, they suffer from two critical limitations: their stochastic reverse processes render entropy intractable (necessitating heuristic exploration), and computing policy gradients through long denoising chains is expensive and unstable. In this work, we show that ODE-based flow matching inherently resolves these issues by enabling both simulation-free policy optimization and tractable entropy computation. Building on this, we introduce Flow Matching Policy Optimization with Mirror Descent and Entropy Constraints (FMER). Our framework exploits this insight in three ways. First, we theoretically establish that minimizing an advantage-weighted conditional flow matching loss acts as a simulation-free surrogate for policy mirror descent. This steers the velocity field toward high-value regions while entirely avoiding backpropagation through the ODE solver. Second, we derive an analytic entropy objective that corrects for the density distortion caused by the $\tanh$ transformation (mapping an unbounded latent space to bounded actions), thereby facilitating principled maximum-entropy optimization. Finally, we dynamically tune the mirror descent temperature based on the effective sample size to enforce a robust trust region during training. Empirical evaluations demonstrate that FMER achieves superior performance on the challenging sparse-reward FrankaKitchen environment, while maintaining competitive results across standard dense-reward MuJoCo benchmarks.

cs.LG↗

arXiv2026

Topological Kalman Filtering on Cell Complexes

Inferring latent dynamics from multivariate time-series defined over topological cell complexes is crucial for capturing the complex, higher-order interactions inherent in real-world systems such as in water, sensor, and transportation networks. However, reconstructing these latent states is challenging because the signals are coupled across higher-order topologies, while high dimensionality, nonlinear observations, and unknown structures increase the difficulty. To address this, we propose a topology-aware state space framework derived from stochastic partial differential equations on cell complexes. State evolution follows heat-like topological diffusion, with perturbations propagating along boundary operators. Under partial observability, we model observations using a cell complex convolution of latent states coupled with a nonlinear mapping. We perform recursive state estimation via an Extended Kalman Filter, simultaneously learning model parameters and uncertainties through an online Expectation-Maximization algorithm. Finally, for scenarios where only lower-order topological structure is known, e.g., nodes and edges, as in critical infrastructure networks, we introduce a heuristic cell identification algorithm to explicitly infer the second-order cell structures. Validations on synthetic and real datasets from water, sensor and transportation networks demonstrate that our approach yields reliable estimates under partial observability and successfully recovers the underlying topological structures.

eess.SP↗

arXiv2026

Learning Product Graphs from Two-dimensional Stationary Signals

Graph learning aims to infer a network structure directly from observed data, enabling the analysis of complex dependencies in irregular domains. Traditional methods focus on scalar signals at each node, ignoring dependencies along additional dimensions such as time, configurations of the observation device, or populations. In this work, we propose a graph signal processing framework for learning graphs from two-dimensional signals, modeled as matrix graph signals generated by joint filtering along both dimensions. This formulation leverages the concept of graph stationarity across the two dimensions and leverages product graph representations to capture structured dependencies. Based on this model, we design an optimization problem that can be solved efficiently and provably recovers the optimal underlying Kronecker/Cartesian/strong product graphs. Experiments on synthetic data demonstrate that our approach achieves higher estimation accuracy and reduced computational cost compared to existing methods.

eess.SP↗

arXiv2026

Stochastic Sequential Decision Making over Expanding Networks with Graph Filtering

Graph filters leverage topological information to process networked data with existing methods mainly studying fixed graphs, ignoring that graphs often expand as nodes continually attach with an unknown pattern. The latter requires developing filter-based decision-making paradigms that take evolution and uncertainty into account. Existing approaches rely on either pre-designed filters or online learning, limited to a myopic view considering only past or present information. To account for future impacts, we propose a stochastic sequential decision-making framework for filtering networked data with a policy that adapts filtering to expanding graphs. By representing filter shifts as agents, we model the filter as a multi-agent system and train the policy following multi-agent reinforcement learning. This accounts for long-term rewards and captures expansion dynamics through sequential decision-making. Moreover, we develop a context-aware graph neural network to parameterize the policy, which tunes filter parameters based on information of both the graph and agents. Experiments on synthetic and real datasets from cold-start recommendation to COVID prediction highlight the benefits of using a sequential decision-making perspective over batch and online filtering alternatives.

cs.LG↗

arXiv2026

Directed Semi-Simplicial Learning with Applications to Brain Activity Decoding

Graph Neural Networks (GNNs) excel at learning from pairwise interactions but often overlook multi-way and hierarchical relationships. Topological Deep Learning (TDL) addresses this limitation by leveraging combinatorial topological spaces. However, existing TDL models are restricted to undirected settings and fail to capture the higher-order directed patterns prevalent in many complex systems, e.g., brain networks, where such interactions are both abundant and functionally significant. To fill this gap, we introduce Semi-Simplicial Neural Networks (SSNs), a principled class of TDL models that operate on semi-simplicial sets -- combinatorial structures that encode directed higher-order motifs and their directional relationships. To enhance scalability, we propose Routing-SSNs, which dynamically select the most informative relations in a learnable manner. We prove that SSNs are strictly more expressive than standard graph and TDL models. We then introduce a new principled framework for brain dynamics representation learning, grounded in the ability of SSNs to provably recover topological descriptors shown to successfully characterize brain activity. Empirically, SSNs achieve state-of-the-art performance on brain dynamics classification tasks, outperforming the second-best model by up to 27%, and message passing GNNs by up to 50% in accuracy. Our results highlight the potential of principled topological models for learning from structured brain data, establishing a unique real-world case study for TDL. We also test SSNs on standard node classification and edge regression tasks, showing competitive performance. We will make the code and data publicly available.

cs.LG↗

arXiv2026

Precision Neural Networks: Joint Graph And Relational Learning

CoVariance Neural Networks (VNNs) perform convolutions on the graph determined by the covariance matrix of the data, which enables expressive and stable covariance-based learning. However, covariance matrices are typically dense, fail to encode conditional independence, and are often precomputed in a task-agnostic way, which may hinder performance. To overcome these limitations, we study Precision Neural Networks (PNNs), i.e., VNNs on the precision matrix - the inverse covariance. The precision matrix naturally encodes statistical independence, often exhibits sparsity, and preserves the covariance spectral structure. To make precision estimation task-aware, we formulate an optimization problem that jointly learns the network parameters and the precision matrix, and solve it via alternating optimization, by sequentially updating the network weights and the precision estimate. We theoretically bound the distance between the estimated and true precision matrices at each iteration, and demonstrate the effectiveness of joint estimation compared to two-step approaches on synthetic and real-world data.

cs.LG↗

arXiv2025

Covariance Scattering Transforms

Machine learning and data processing techniques relying on covariance information are widespread as they identify meaningful patterns in unsupervised and unlabeled settings. As a prominent example, Principal Component Analysis (PCA) projects data points onto the eigenvectors of their covariance matrix, capturing the directions of maximum variance. This mapping, however, falls short in two directions: it fails to capture information in low-variance directions, relevant when, e.g., the data contains high-variance noise; and it provides unstable results in low-sample regimes, especially when covariance eigenvalues are close. CoVariance Neural Networks (VNNs), i.e., graph neural networks using the covariance matrix as a graph, show improved stability to estimation errors and learn more expressive functions in the covariance spectrum than PCA, but require training and operate in a labeled setup. To get the benefits of both worlds, we propose Covariance Scattering Transforms (CSTs), deep untrained networks that sequentially apply filters localized in the covariance spectrum to the input data and produce expressive hierarchical representations via nonlinearities. We define the filters as covariance wavelets that capture specific and detailed covariance spectral patterns. We improve CSTs' computational and memory efficiency via a pruning mechanism, and we prove that their error due to finite-sample covariance estimations is less sensitive to close covariance eigenvalues compared to PCA, improving their stability. Our experiments on age prediction from cortical thickness measurements on 4 datasets collecting patients with neurodegenerative diseases show that CSTs produce stable representations in low-data settings, as VNNs but without any training, and lead to comparable or better predictions w.r.t. more complex learning models.

cs.LG↗

arXiv2025

Graph-Aware Diffusion for Signal Generation

We study the problem of generating graph signals from unknown distributions defined over given graphs, relevant to domains such as recommender systems or sensor networks. Our approach builds on generative diffusion models, which are well established in vision and graph generation but remain underexplored for graph signals. Existing methods lack generality, either ignoring the graph structure in the forward process or designing graph-aware mechanisms tailored to specific domains. We adopt a forward process that incorporates the graph through the heat equation. Rather than relying on the standard formulation, we consider a time-warped coefficient to mitigate the exponential decay of the drift term, yielding a graph-aware generative diffusion model (GAD). We analyze its forward dynamics, proving convergence to a Gaussian Markov random field with covariance parametrized by the graph Laplacian, and interpret the backward dynamics as a sequence of graph-signal denoising problems. Finally, we demonstrate the advantages of GAD on synthetic data, real traffic speed measurements, and a temperature sensor network.

cs.LG↗

arXiv2025

CoVariance Filters and Neural Networks over Hilbert Spaces

CoVariance Neural Networks (VNNs) perform graph convolutions on the empirical covariance matrix of signals defined over finite-dimensional Hilbert spaces, motivated by robustness and transferability properties. Yet, little is known about how these arguments extend to infinite-dimensional Hilbert spaces. In this work, we take a first step by introducing a novel convolutional learning framework for signals defined over infinite-dimensional Hilbert spaces, centered on the (empirical) covariance operator. We constructively define Hilbert coVariance Filters (HVFs) and design Hilbert coVariance Networks (HVNs) as stacks of HVF filterbanks with nonlinear activations. We propose a principled discretization procedure, and we prove that empirical HVFs can recover the Functional PCA (FPCA) of the filtered signals. We then describe the versatility of our framework with examples ranging from multivariate real-valued functions to reproducing kernel Hilbert spaces. Finally, we validate HVNs on both synthetic and real-world time-series classification tasks, showing robust performance compared to MLP and FPCA-based classifiers.

cs.LG↗

arXiv2025

GKNet: Graph Kalman Filtering and Model Inference via Model-based Deep Learning

Inference tasks with time series over graphs are of importance in applications such as urban water networks, economics, and networked neuroscience. Addressing these tasks typically relies on identifying a computationally affordable model that jointly captures the graph-temporal patterns of the data. In this work, we propose a graph-aware state space model for graph time series, where both the latent state and the observation equation are parametric graph-induced models with a limited number of parameters that need to be learned. More specifically, we consider the state equation to follow a stochastic partial differential equation driven by noise over the graphs edges accounting not only for potential edge uncertainties but also for increasing the degrees of freedom in the latter in a tractable manner. The graph structure conditioning of the noise dispersion allows the state variable to deviate from the stochastic process in certain neighborhoods. The observation model is a sampled and graph-filtered version of the state capturing multi-hop neighboring influence. The goal is to learn the parameters in both state and observation models from the partially observed data for downstream tasks such as prediction and imputation. The model is inferred first through a maximum likelihood approach that provides theoretical tractability but is limited in expressivity and scalability. To improve on the latter, we use the state-space formulation to build a principled deep learning architecture that jointly learns the parameters and tracks the state in an end-to-end manner in the spirit of Kalman neural networks.

cs.LG↗

arXiv2025

Graph signal aware decomposition of dynamic networks via latent graphs

Dynamics on and of networks refer to changes in topology and node-associated signals, respectively and are pervasive in many socio-technological systems, including social, biological, and infrastructure networks. Due to practical constraints, privacy concerns, or malfunctions, we often observe only a fraction of the topological evolution and associated signal, which not only hinders downstream tasks but also restricts our analysis of network evolution. Such aspects could be mitigated by moving our attention at the underlying latent driving factors of the network evolution, which can be naturally uncovered via low-rank tensor decomposition. Tensor-based methods provide a powerful means of uncovering the underlying factors of network evolution through low-rank decompositions. However, the extracted embeddings typically lack a relational structure and are obtained independently from the node signals. This disconnect reduces the interpretability of the embeddings and overlooks the coupling between topology and signals. To address these limitations, we propose a novel two-way decomposition to represent a dynamic graph topology, where the structural evolution is captured by a linear combination of latent graph adjacency matrices reflecting the overall joint evolution of both the topology and the signal. Using spatio-temporal data, we estimate the latent adjacency matrices and their temporal scaling signatures via alternating minimization, and prove that our approach converges to a stationary point. Numerical results show that the proposed method recovers individually and collectively expressive latent graphs, outperforming both standard tensor-based decompositions and signal-based topology identification methods in reconstructing the missing network especially when observations are limited.

eess.SP↗

arXiv2025

Matched Topological Subspace Detector

Topological spaces, represented by simplicial complexes, capture richer relationships than graphs by modeling interactions not only between nodes but also among higher-order entities, such as edges or triangles. This motivates the representation of information defined in irregular domains as topological signals. By leveraging the spectral dualities of Hodge and Dirac theory, practical topological signals often concentrate in specific spectral subspaces (e.g., gradient or curl). For instance, in a foreign currency exchange network, the exchange flow signals typically satisfy the arbitrage-free condition and hence are curl-free. However, the presence of anomalies can disrupt these conditions, causing the signals to deviate from such subspaces. In this work, we formulate a hypothesis testing framework to detect whether simplicial complex signals lie in specific subspaces in a principled and tractable manner. Concretely, we propose Neyman-Pearson matched topological subspace detectors for signals defined at a single simplicial level (such as edges) or jointly across all levels of a simplicial complex. The (energy-based projection) proposed detectors handle missing values, provide closed-form performance analysis, and effectively capture the unique topological properties of the data. We demonstrate the effectiveness of the proposed topological detectors on various real-world data, including foreign currency exchange networks.

stat.ML↗

arXiv2025

Towards Carbon Footprint-Aware Recommender Systems for Greener Item Recommendation

The commodity and widespread use of online shopping are having an unprecedented impact on climate, with emission figures from key actors that are easily comparable to those of a large-scale metropolis. Despite online shopping being fueled by recommender systems (RecSys) algorithms, the role and potential of the latter in promoting more sustainable choices is little studied. One of the main reasons for this could be attributed to the lack of a dataset containing carbon footprint emissions for the items. While building such a dataset is a rather challenging task, its presence is pivotal for opening the doors to novel perspectives, evaluations, and methods for RecSys research. In this paper, we target this bottleneck and study the environmental role of RecSys algorithms. First, we mine a dataset that includes carbon footprint emissions for its items. Then, we benchmark conventional RecSys algorithms in terms of accuracy and sustainability as two faces of the same coin. We find that RecSys algorithms optimized for accuracy overlook greenness and that longer recommendation lists are greener but less accurate. Then, we show that a simple reranking approach that accounts for the item's carbon footprint can establish a better trade-off between accuracy and greenness. This reranking approach is modular, ready to use, and can be applied to any RecSys algorithm without the need to alter the underlying mechanisms or retrain models. Our results show that a small sacrifice of accuracy can lead to significant improvements of recommendation greenness across all algorithms and list lengths. Arguably, this accuracy-greenness trade-off could even be seen as an enhancement of user satisfaction, particularly for purpose-driven users who prioritize the environmental impact of their choices. We anticipate this work will serve as the starting point for studying RecSys for more sustainable recommendations.

cs.IR↗