arXiv subjects

Francisco Silva

Publications and source records attributed to Francisco Silva.

12 recordsLinked to original sources

arXiv2026

Eikonal quasinormal modes, greybody factors and shadow of charged accelerating black holes

We show that the quasinormal modes, in the eikonal limit, for accelerating (non-rotating) black holes, are related to the angular velocity of the circular null geodesics and to the corresponding Lyapunov exponent, in the same way as the ones for spherically symmetric black holes are. We compute those quasinormal modes and greybody factors for neutral and charged accelerating black holes, considering massless test scalar fields, and we show that the results are universal for perturbations of any spin. We also determine the radius of the shadow cast by these black holes. Our results for charged black holes are valid for the Reissner-Nordstrom solution simply by setting the acceleration to zero.

gr-qc↗

arXiv2026

Gradient extremals, talwegs, valleys, and directional alignment for generic gradient descent

Gradient extremals are loci along which the gradient is an eigenvector of the Hessian. These objects provide a natural geometric framework connecting several notions, notably valleys and talwegs, which we analyze from a variational viewpoint in the generic case. We then show that trajectories of the gradient flow and of its discrete counterpart exhibit directional alignment with the tangent spaces to gradient extremals, and generically to the talweg. Under non-resonance assumptions, and in contrast with the quadratic case, alignment rates are governed either by the first spectral gap or by the smallest eigenvalue of the Hessian at the limit point. Nonlinearities and the step length may both distort these rates in a complex manner. We further prove a volume concentration phenomenon emphasizing the structuring role of gradient extremals: for large times, the images of sets of initial conditions concentrate inside valleys and asymptotically around talwegs.

math.OC↗

arXiv2023

In Search of Deep Learning Architectures for Load Forecasting: A Comparative Analysis and the Impact of the Covid-19 Pandemic on Model Performance

In power grids, short-term load forecasting (STLF) is crucial as it contributes to the optimization of their reliability, emissions, and costs, while it enables the participation of energy companies in the energy market. STLF is a challenging task, due to the complex demand of active and reactive power from multiple types of electrical loads and their dependence on numerous exogenous variables. Amongst them, special circumstances, such as the COVID-19 pandemic, can often be the reason behind distribution shifts of load series. This work conducts a comparative study of Deep Learning (DL) architectures, namely Neural Basis Expansion Analysis Time Series Forecasting (N-BEATS), Long Short-Term Memory (LSTM), and Temporal Convolutional Networks (TCN), with respect to forecasting accuracy and training sustainability, meanwhile examining their out-of-distribution generalization capabilities during the COVID-19 pandemic era. A Pattern Sequence Forecasting (PSF) model is used as baseline. The case study focuses on day-ahead forecasts for the Portuguese national 15-minute resolution net load time series. The results can be leveraged by energy companies and network operators (i) to reinforce their forecasting toolkit with state-of-the-art DL models; (ii) to become aware of the serious consequences of crisis events on model performance; (iii) as a high-level model evaluation, deployment, and sustainability guide within a smart grid context.

cs.LG↗

arXiv2020

Single-shot d-scan technique for ultrashort laser pulse characterization using transverse second-harmonic generation in random nonlinear crystals

We demonstrate a novel dispersion-scan (d-scan) scheme for single-shot temporal characterization of ultrashort laser pulses. The novelty of this method relies on the use of a highly dispersive crystal featuring antiparallel nonlinear domains with a random distribution and size. This crystal, capable of generating a transverse second-harmonic signal, acts simultaneously as the dispersive element and the nonlinear medium of the d-scan device. The resulting in-line architecture makes the technique very simple and robust, allowing the acquisition of single-shot d-scan traces in real time. In addition, the technique can be further simplified by avoiding the need of dispersion pre-compensation. The retrieved pulses are in very good agreement with independent FROG measurements. We also apply the new single-shot d-scan to a TW-class laser equipped with a programmable pulse shaper, obtaining an excellent agreement between the applied and the d-scan retrieved dispersions.

physics.optics↗

arXiv2018

Perceptrons from Memristors

Memristors, resistors with memory whose outputs depend on the history of their inputs, have been used with success in neuromorphic architectures, particularly as synapses and non-volatile memories. However, to the best of our knowledge, no model for a network in which both the synapses and the neurons are implemented using memristors has been proposed so far. In the present work we introduce models for single and multilayer perceptrons based exclusively on memristors. We adapt the delta rule to the memristor-based single-layer perceptron and the backpropagation algorithm to the memristor-based multilayer perceptron. Our results show that both perform as expected for perceptrons, including satisfying Minsky-Papert's theorem. As a consequence of the Universal Approximation Theorem, they also show that memristors are universal function approximators. By using memristors for both the neurons and the synapses, our models pave the way for novel memristor-based neural network architectures and algorithms. A neural network based on memristors could show advantages in terms of energy conservation and open up possibilities for other learning systems to be adapted to a memristor-based paradigm, both in the classical and quantum learning realms.

cs.ET↗

arXiv2017

Universal route to optimal few- to single-cycle pulse generation in hollow-core fiber compressors

Gas-filled hollow-core fiber (HCF) pulse post-compressors generating few- to single-cycle pulses are a key enabling tool for attosecond science and ultrafast spectroscopy. Achieving optimum performance in this regime can be extremely challenging due to the ultra-broad bandwidth of the pulses and the need of an adequate temporal diagnostic. These difficulties have hindered the full exploitation of HCF post-compressors, namely the generation of stable and high-quality near-Fourier-transform-limited pulses. Here we show that, independently of conditions such as the type of gas or the laser system used, there is a universal route to obtain the shortest stable output pulse down to the single-cycle regime. Numerical simulations and experimental measurements performed with the dispersion-scan technique reveal that, in quite general conditions, post-compressed pulses exhibit a residual third-order dispersion intrinsic to optimum nonlinear propagation within the fiber, in agreement with measurements independently performed in several laboratories around the world. The understanding of this effect and its adequate correction, e.g. using simple transparent optical media, enables achieving high-quality post-compressed pulses with only minor changes in existing setups. These optimized sources have impact in many fields of science and technology and should enable new and exciting applications in the few- to single-cycle pulse regime.

physics.optics↗

arXiv2017

Tunable high-harmonic generation by chromatic focusing of few-cycle laser pulses

In this work we study the impact of chromatic focusing of few-cycle laser pulses on high-order harmonic generation (HHG) through analysis of the emitted extreme ultraviolet (XUV) radiation. Chromatic focusing is usually avoided in the few-cycle regime, as the pulse spatio-temporal structure may be highly distorted by the spatiotemporal aberrations. Here, however, we demonstrate it as an additional control parameter to modify the generated XUV radiation. We present experiments where few-cycle pulses are focused by a singlet lens in a Kr gas jet. The chromatic distribution of focal lengths allows us to tune HHG spectra by changing the relative singlet-target distance. Interestingly, we also show that the degree of chromatic aberration needed to this control does not degrade substantially the harmonic conversion efficiency, still allowing for the generation of supercontinua with the chirped-pulse scheme, demonstrated previously for achromatic focussing. We back up our experiments with theoretical simulations reproducing the experimental HHG results depending on diverse parameters (input pulse spectral phase, pulse duration, focus position) and proving that, under the considered parameters, the attosecond pulse train remains very similar to the achromatic case, even showing cases of isolated attosecond pulse generation for near single-cycle driving pulses.

physics.optics↗

arXiv2017

Petahertz optical response in graphene

The temporal dynamics of charge carriers determines the speed with which electronics can be realized in condensed matter, and their direct manipulation with optical fields promises electronic processing at unprecedented petahertz frequencies, consisting in a million-fold increase from state of the art technology. Graphene is of particular interest for the implementation of petahertz optoelectronics due to its unique transport properties, such as high carrier mobility with near-ballistic transport and exceptionally strong coupling to optical fields. The back action of carriers in response to an optical field is therefore of key importance towards applications. Here we investigate the instantaneous response of graphene to petahertz optical fields and elucidate the role of hot carriers on a sub-100 fs timescale. Measurements of the nonlinear response and its dependence on interaction time and field polarization allow us to identify the back action of hot carriers over timescales that are commensurate with the optical field. An intuitive picture is given for the carrier trajectories in response to the optical-field polarization state. We note that the peculiar interplay between optical fields and charge carriers in graphene may also apply to surface states in topological insulators with similar Dirac cone dispersion relations.

cond-mat.mes-hall↗

arXiv2017

Sensitivity analysis for expected utility maximization in incomplete Brownian market models

We examine the issue of sensitivity with respect to model parameters for the problem of utility maximization from final wealth in an incomplete Samuelson model and mainly, but not exclusively, for utility functions of positive power-type. The method consists in moving the parameters through change of measure, which we call a weak perturbation, decoupling the usual wealth equation from the varying parameters. By rewriting the maximization problem in terms of a convex-analytical support function of a weakly-compact set, crucially leveraging on the work of Backhoff and Fontbona (SIFIN 2016), the previous formulation let us prove the Hadamard directional differentiability of the value function w.r.t. the drift and interest rate parameters, as well as for volatility matrices under a stability condition on their Kernel, and derive explicit expressions for the directional derivatives. We contrast our proposed weak perturbations against what we call strong perturbations, where the wealth equation is directly influenced by the changing parameters. Contrary to conventional wisdom, we find that both points of view generally yield different sensitivities unless e.g. if initial parameters and their perturbations are deterministic.

q-fin.MF↗

arXiv2015

Single-shot implementation of dispersion-scan for the characterization of ultrashort laser pulses

We demonstrate a novel, single-shot ultrafast diagnostic, based on the dispersion-scan (d-scan) technique. In this implementation, rather than scanning wedges to vary the dispersion as in standard d-scan, the pulse to be measured experiences a spatially varying amount of dispersion in a Littrow prism. The resulting beam is then imaged into a second-harmonic generation crystal and an imaging spectrometer is used to measure the two-dimensional trace, which is analyzed using the d-scan retrieval algorithm. We compare the single-shot implementation with the standard d-scan for the measurement of sub-3.5-fs pulses from a hollow core fiber pulse compressor. We show that the retrieval algorithm used to extract amplitude and phase of the pulse provides comparable results, proving the validity of the new single-shot implementation down to near single-cycle durations.

physics.optics↗

arXiv2014

Spatio-temporal isolation of attosecond soft X-ray pulses in the water window

We demonstrate experimentally the isolation of single attosecond pulses at the carbon K- shell edge in the soft-X-ray water window. Attosecond pulses at photon energies that cover the principal absorption edges of the building blocks of materials are a prerequisite for time resolved probing of the triggering events leading to electronic dynamics such as exciton formation and annihilation. Herewith, we demonstrate successful isolation of individual attosecond pulses at the carbon K edge (284 eV) with a pulse duration below 400 as and with a bandwidth supporting a 30 as pulse duration. Our approach is based on spatio-temporal isolation of ponderomotively shifted harmonics and validates a straightforward and scalable approach for robust and reproducible attosecond pulse isolation.

physics.optics↗

arXiv2014

Some sensitivity results in stochastic optimal control: A Lagrange multiplier point of view

In this work we provide a first order sensitivity analysis of some parameterized stochastic optimal control problems. The parameters can be given by random processes. The main tool is the one-to-one correspondence between the adjoint states appearing in a weak form of the stochastic Pontryagin principle and the Lagrange multipliers associated to the state equation.

math.OC↗