Santiago Hernandez | Searcharxiv

arXiv2025

Shedding Light on Naked Singularities

Electromagnetic waves propagating in the background provided by a spacetime hosting a strong curvature, naked singularity, are fully studied. The analysis is performed not only in the realm of geometrical optics -- which, not surprisingly, proves to be inadequate in the strong field regime characterizing the vicinity of the singularity -- but also in the physical one in which the field amplitudes must necessarily be incorporated into the description. In addition to the expected divergent outcome with regard to the field amplitudes and power flux as the waves approach the singularity, we found a number of regular (bounded) solutions which seem to coexist with the unbridled effects of the spacetime curvature. In some of them, the singularity operates as a perfect mirror reflecting the surrounding fields. Strikingly, other solutions exhibit a perfectly well behaved, bounded power flux as they propagate towards the singularity, suggesting thus the possibility of having electromagnetic energy transference through it.

gr-qc↗

arXiv2023

BeCAPTCHA-Type: Biometric Keystroke Data Generation for Improved Bot Detection

This work proposes a data driven learning model for the synthesis of keystroke biometric data. The proposed method is compared with two statistical approaches based on Universal and User-dependent models. These approaches are validated on the bot detection task, using the keystroke synthetic data to improve the training process of keystroke-based bot detection systems. Our experimental framework considers a dataset with 136 million keystroke events from 168 thousand subjects. We have analyzed the performance of the three synthesis approaches through qualitative and quantitative experiments. Different bot detectors are considered based on several supervised classifiers (Support Vector Machine, Random Forest, Gaussian Naive Bayes and a Long Short-Term Memory network) and a learning framework including human and synthetic samples. The experiments demonstrate the realism of the synthetic samples. The classification results suggest that in scenarios with large labeled data, these synthetic samples can be detected with high accuracy. However, in few-shot learning scenarios it represents an important challenge. Furthermore, these results show the great potential of the presented models.

cs.LG↗