Source author record

Viktoria Fodor

Viktoria Fodor appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

Machine Learning Networking and Internet Architecture Information Theory math.IT Applications Artificial Intelligence eess.SP Emerging Technologies Methodology Performance physics.soc-ph

Catalog footprint

What is connected

7works

11topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2026arXiv

Deep Neural Sheaf Diffusion

Deep Graph Neural Networks (GNNs) are essential for capturing complex dependencies in graph-structured data. However, scaling GNNs to depth remains challenging, as stacking layers leads to representation collapse and diminishing sensitivity due to repeated aggregation. While Neural Sheaf Diffusion (NSD) provides strong theoretical guarantees against such collapse, these guarantees do not translate to practice: as depth increases, the disagreement signal of the sheaf Laplacian vanishes, limiting the contribution of deeper layers. We identify mechanisms that hinder NSD effectiveness at depth and propose \emph{Deep Neural Sheaf Diffusion} (DNSD), which replaces the sheaf Laplacian with a sheaf adjacency operator to maintain informative signals across layers. This is complemented by normalization, odd nonlinearities, and gating. To provide a principled explanation of the expected performance improvement, we contrast sheaf diffusion to graph attention mechanisms, highlighting that DNSD replaces scalar attention scores with matrix-valued edge functions and normalizes node representations rather than attention scores. We demonstrate empirically that DNSD effectively utilizes deep aggregation in graph tasks, outperforming GNN and NSD baselines with up to 30pp accuracy on synthetic long-range datasets, and consistently outperforming them on real-world benchmarks. These results position sheaf-based architectures as a promising building block for graph foundation models by supporting effective deep architectures.

preprint2024arXiv

Hierarchical Over-the-Air Federated Learning with Awareness of Interference and Data Heterogeneity

When implementing hierarchical federated learning over wireless networks, scalability assurance and the ability to handle both interference and device data heterogeneity are crucial. This work introduces a learning method designed to address these challenges, along with a scalable transmission scheme that efficiently uses a single wireless resource through over-the-air computation. To provide resistance against data heterogeneity, we employ gradient aggregations. Meanwhile, the impact of interference is minimized through optimized receiver normalizing factors. For this, we model a multi-cluster wireless network using stochastic geometry, and characterize the mean squared error of the aggregation estimations as a function of the network parameters. We show that despite the interference and the data heterogeneity, the proposed scheme achieves high learning accuracy and can significantly outperform the conventional hierarchical algorithm.

preprint2022arXiv

Wireless for Machine Learning

As data generation increasingly takes place on devices without a wired connection, machine learning (ML) related traffic will be ubiquitous in wireless networks. Many studies have shown that traditional wireless protocols are highly inefficient or unsustainable to support ML, which creates the need for new wireless communication methods. In this survey, we give an exhaustive review of the state-of-the-art wireless methods that are specifically designed to support ML services over distributed datasets. Currently, there are two clear themes within the literature, analog over-the-air computation and digital radio resource management optimized for ML. This survey gives a comprehensive introduction to these methods, reviews the most important works, highlights open problems, and discusses application scenarios.

preprint2020arXiv

Will Scale-free Popularity Develop Scale-free Geo-social Networks?

Empirical results show that spatial factors such as distance, population density and communication range affect our social activities, also reflected by the development of ties in social networks. This motivates the need for social network models that take these spatial factors into account. Therefore, in this paper we propose a gravity-low-based geo-social network model, where connections develop according to the popularity of the individuals, but are constrained through their geographic distance and the surrounding population density. Specifically, we consider a power-law distributed popularity, and random node positions governed by a Poisson point process. We evaluate the characteristics of the emerging networks, considering the degree distribution, the average degree of neighbors and the local clustering coefficient. These local metrics reflect the robustness of the network, the information dissemination speed and the communication locality. We show that unless the communication range is strictly limited, the emerging networks are scale-free, with a rank exponent affected by the spatial factors. Even the average neighbor degree and the local clustering coefficient show tendencies known in non-geographic scale-free networks, at least when considering individuals with low popularity. At high-popularity values, however, the spatial constraints lead to popularity-independent average neighbor degrees and clustering coefficients.

preprint2015arXiv

Green Sensing and Access: Energy-Throughput Tradeoffs in Cognitive Networking

Limited spectrum resources and dramatic growth of high data rate applications have motivated opportunistic spectrum access utilizing the promising concept of cognitive networks. Although this concept has emerged primarily to enhance spectrum utilization and to allow the coexistence of heterogeneous network technologies, the importance of energy consumption imposes additional challenges, because energy consumption and communication performance can be at odds. In this paper, the approaches for energy efficient spectrum sensing and spectrum handoff, fundamental building blocks of cognitive networks is investigated. The tradeoff between energy consumption and throughput, under local as well as under cooperative sensing are characterized, and what further aspects need to be investigated to achieve energy efficient cognitive operation under various application requirements are discussed.

preprint2013arXiv

Energy Efficient Spectrum Sensing and Handoff Strategies in Cognitive Radio Networks

The limited spectrum resources and dramatic growth of high data rate communications have motivated opportunistic spectrum access using the promising concept of cognitive radio networks. Although this concept has emerged primarily to enhance spectrum utilization, the importance of energy consumption poses new challenges, because energy efficiency and communication performance can be at odds. In this paper, the existing approaches to energy efficiency spectrum sensing and handoff are classified. The tradeoff between energy consumption and throughput is established as function of the numerous design parameters of cognitive radio networks, both in the case of local and of cooperative spectrum sensing. It is argued that a number of important aspects still needs to be researched, such as fairness, dynamic behavior, reactive and proactive schemes for energy efficiency.

preprint2009arXiv

Heterogeneous attachment strategies optimize the topology of dynamic wireless networks

In optimizing the topology of wireless networks built of a dynamic set of spatially embedded agents, there are many trade-offs to be dealt with. The network should preferably be as small (in the sense that the average, or maximal, pathlength is short) as possible, it should be robust to failures, not consume too much power, and so on. In this paper, we investigate simple models of how agents can choose their neighbors in such an environment. In our model of attachment, we can tune from one situation where agents prefer to attach to others in closest proximity, to a situation where distance is ignored (and thus attachments can be made to agents further away). We evaluate this scenario with several performance measures and find that the optimal topologies, for most of the quantities, is obtained for strategies resulting in a mix of most local and a few random connections.