Source author record

Filip Svoboda

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

ResearcherUnclaimed source record

Computer Vision Cryptography and Security eess.IV Machine Learning physics.data-an physics.geo-ph physics.space-ph

Catalog footprint

What is connected

3works

7topics

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

Attacks on fairness in Federated Learning

Federated Learning is an important emerging distributed training paradigm that keeps data private on clients. It is now well understood that by controlling only a small subset of FL clients, it is possible to introduce a backdoor to a federated learning model, in the presence of certain attributes. In this paper, we present a new type of attack that compromises the fairness of the trained model. Fairness is understood to be the attribute-level performance distribution of a trained model. It is particularly salient in domains where, for example, skewed accuracy discrimination between subpopulations could have disastrous consequences. We find that by employing a threat model similar to that of a backdoor attack, an attacker is able to influence the aggregated model to have an unfair performance distribution between any given set of attributes. Furthermore, we find that this attack is possible by controlling only a single client. While combating naturally induced unfairness in FL has previously been discussed in depth, its artificially induced kind has been neglected. We show that defending against attacks on fairness should be a critical consideration in any situation where unfairness in a trained model could benefit a user who participated in its training.

preprint2026arXiv

deSEO: Physics-Aware Dataset Creation for High-Resolution Satellite Image Shadow Removal

Shadows cast by terrain and tall structures remain a major obstacle for high-resolution satellite image analysis, degrading classification, detection, and 3D reconstruction performance. Public resources offering geometry-consistent paired shadow/shadow-free satellite imagery are essentially missing, and most Earth-observation datasets are designed for shadow detection or 3D modelling rather than removal. Existing deep shadow-removal datasets either target ground-level or aerial scenes or rely on unpaired and weakly supervised formulations rather than explicit satellite pairs. We address this gap with deSEO, a geometry-aware and physics-informed methodology that, to the best of our knowledge, is the first to derive paired supervision for satellite shadow removal from the S-EO shadow detection dataset through a fully replicable pipeline. For each tile, deSEO selects a minimally shadowed acquisition as a weak reference and pairs it with shadowed counterparts using temporal and geometric filtering, Jacobian-based orientation normalisation, and LoFTR-RANSAC registration. A per-pixel validity mask restricts learning to reliably aligned regions, enabling supervision despite residual off-nadir parallax. In addition to this paired dataset, we develop a DSM-aware deshadowing model that combines residual translation, perceptual objectives, and mask-constrained adversarial learning. In contrast, a direct adaptation of a UAV-based SRNet/pix2pix architecture fails to converge under satellite viewpoint variability. Our model consistently reduces the visual impact of cast shadows across diverse illumination and viewing conditions, achieving improved structural and perceptual fidelity on held-out scenes. deSEO therefore provides the first reproducible, geometry-aware paired dataset and baseline for shadow removal in satellite Earth observation.

preprint2024arXiv

A Global Analysis of Pre-Earthquake Ionospheric Anomalies

Local ionospheric density anomalies have been reported in the days prior to major earthquakes. This global study statistically investigates whether consistent ionospheric anomalies occur in the 24 hours prior to earthquakes across different regions, magnitudes, temporal and spatial scales. We match earthquake data to Total Electron Content (TEC) data from 2000-2020 at a higher resolution and cadence than previous assessed. Globally, no significant, consistent anomaly is found. Regionally, statistically significant ionospheric anomalies arise in the 12 hours prior to earthquakes with $p \leq 0.01$ following Wilcoxon tests. For the Japanese region we find a median negative ionospheric anomaly of around 0.5 TECU between 3 and 8 hours before earthquakes. For the South American region, the median TEC is enhanced by up to ~ 2 TECU, between 7 and 10 hours before an event. We show that the results are robust to different definitions of the ''local'' region and earthquake magnitude. This demonstrates the promise of monitoring the ionosphere as part of a multimodal earthquake forecasting system.