Machine learning methods, such as deep learning, show promising results in the medical domain. However, the lack of interpretability of these algorithms may hinder their applicability to medical decision support systems. This paper studies an interpr
etable deep learning technique, called SincNet. SincNet is a convolutional neural network that efficiently learns customized band-pass filters through trainable sinc-functions. In this study, we use SincNet to analyze the neural activity of individuals with Autism Spectrum Disorder (ASD), who experience characteristic differences in neural oscillatory activity. In particular, we propose a novel SincNet-based neural network for detecting emotions in ASD patients using EEG signals. The learned filters can be easily inspected to detect which part of the EEG spectrum is used for predicting emotions. We found that our system automatically learns the high-$alpha$ (9-13 Hz) and $beta$ (13-30 Hz) band suppression often present in individuals with ASD. This result is consistent with recent neuroscience studies on emotion recognition, which found an association between these band suppressions and the behavioral deficits observed in individuals with ASD. The improved interpretability of SincNet is achieved without sacrificing performance in emotion recognition.
In this paper we revisit the problem of identifying bona fide cluster Cepheids by performing an all-sky search for Cepheids associated with open clusters and making use of state-of-the-art catalogued information for both Cepheids and clusters, based
on the unparalleled astrometric precision of the second and early third data releases of the Gaia satellite. We determine membership probabilities by following a Bayesian approach using spatial and kinematic information of the potential cluster-Cepheid pairs. We confirm 19 Cepheid-cluster associations considered in previous studies as bona-fide, and question the established cluster membership of six other associations. In addition, we identify 138 cluster Cepheid candidates of potential interest, mostly in recently discovered open clusters. We report on at least two new clusters possibly hosting more than one Cepheid. Furthermore, we explore the feasibility of using open clusters hosting Cepheids to empirically determine the Cepheid period-age relation through the use of Gaia and 2MASS photometry and a semi-automated method to derive cluster ages. We conclude that the usage of cluster Cepheids as tentative probes of the period-age relations still faces difficulties due to the sparsely populated red giant branch and the stochastically sampled main-sequence turn-off of the open clusters, making age determinations a challenging task. This biases the age-dateable cluster selection for Cepheid period-age studies towards older and high-mass clusters.
Can satellite data be used to address challenges currently faced by the Offshore Renewable Energy (ORE) sector? What benefit can satellite observations bring to resource assessment and maintenance of ORE farms? Can satellite observations be used to a
ssess the environmental impact of offshore renewables leading towards a more sustainable ORE sector? This review paper faces these questions presenting a holistic view of the current interactions between satellite and ORE sectors, and future needs to make this partnerships grow. The aim of the work is to start the conversation between these sectors by establishing a common ground. We present offshore needs and satellite technology limitations, as well as potential opportunities and areas of growth. To better understand this, the reader is guided through the history, current developments, challenges and future of offshore wind, tidal and wave energy technologies. Then, an overview on satellite observations for ocean applications is given, covering types of instruments and how they are used to provide different metocean variables, satellite performance, and data processing and integration. Past, present and future satellite missions are also discussed. Finally, the paper focuses on innovation opportunities and the potential of synergies between the ORE and satellite sectors. Specifically, we pay attention to improvements that satellite observations could bring to standard measurement techniques: assessing uncertainty, wind, tidal and wave conditions forecast, as well as environmental monitoring from space. Satellite-enabled measurement of ocean physical processes and applications for fisheries, mammals and birds, and habitat change, are also discussed in depth.
The discovery of very distant stars in the halo of the Milky Way provides valuable tracers on the Milky Way mass and its formation. Beyond 100 kpc from the Galactic center, most of the stars are likely to be in faint dwarf galaxies or tidal debris fr
om recently accreted dwarfs, making the outer reaches of the Galaxy important for understanding the Milky Ways accretion history. However, distant stars in the halo are scarce. In that context, RR Lyrae are ideal probes of the distant halo as they are intrinsically bright and thus can be seen at large distances, follow well-known period-luminosity relations that enable precise distance measurements, and are easily identifiable in time-series data. Therefore, a detailed study of RR Lyrae will help us understand the accreted outskirts of the Milky Way. In this contribution, we present the current state of our systematic search for distant RR Lyrae stars in the halo using the DECam imager at the 4m telescope on Cerro Tololo (Chile). The total surveyed area consists of more than 110 DECam fields (~ 350 sq. deg) and includes two recent independent campaigns carried out in 2017 and 2018 with which we have detected > 650 candidate RR Lyrae stars. Here we describe the methodology followed to analyze the two latest campaigns. Our catalog contains a considerable number of candidate RR Lyrae beyond 100 kpc, and reaches out up to ~ 250 kpc. The number of distant RR Lyrae found is consistent with recent studies of the outer halo. These stars provide a set of important probes of the mass of the Milky Way, the nature of the halo, and the accretion history of the Galactic outskirts.
Classical Cepheids in open clusters are key ingredients for stellar population studies and the characterization of variable stars, as they are tracers of young and massive populations and of recent star formation episodes. Cluster Cepheids are of par
ticular importance since they can be age dated by using the clusters stellar population to obtain the Cepheid period-luminosity-age relation. In this contribution, we present the preliminary results of an all-sky search for classical Cepheids in Galactic open clusters by taking advantage of the unprecedented astrometric precision of the second data release of the Gaia satellite. To do this, we determined membership probabilities by performing a Bayesian analysis based on the spatial distribution of Cepheids and clusters, and their kinematics. Here we describe our adopted methodology.
The ubiquitous relativistic jet phenomena associated with black holes play a major role in high and very-high-energy (VHE) astrophysics. In particular, observations have demonstrated that blazars show VHE emission with time-variability from days (in
the GeV band) to minutes (in the TeV band), implying very compact emission regions. The real mechanism able to explain the particle acceleration process responsible for this emission is still debated, but magnetic reconnection has been lately discussed as a strong potential candidate and, in some circumstances, as the only possible one. In this work, we present the results of three-dimensional special relativistic magnetohydrodynamic simulations of the development of reconnection events driven by turbulence induced by current-driven kink instability along a relativistic jet. We have performed a systematic identification of all reconnection regions in the system, characterizing their local magnetic field topology and quantifying the reconnection rates. We obtained average rates of $0.051pm0.026$ (in units of the Alfv{e}n speed) which are comparable to the predictions of the theory of turbulence-induced fast reconnection. Detailed statistical analysis also demonstrated that the fast reconnection events follow a log-normal distribution, which is a signature of its turbulent origin. To probe the robustness of our method, we have applied our results to the blazar Mrk 421. Building a synthetic light curve from the integrated power of the magnetic reconnection events, we evaluated the time-variability from a power spectral density analysis, obtaining a good agreement with the observations in the GeV band. This suggests that turbulent fast magnetic reconnection driven by kink instability can be a possible process behind the high energy emission variability phenomena observed in blazars.
Particle acceleration in magnetized relativistic jets still puzzles theorists, specially when one tries to explain the highly variable emission observed in blazar jets or gamma-ray bursts putting severe constraints on current models. In this work we
investigate the acceleration of particles injected in a three-dimensional relativistic magnetohydrodynamical jet subject to current driven kink instability (CDKI), which drives turbulence and fast magnetic reconnection. Test protons injected in the nearly stationary snapshots of the jet, experience an exponential acceleration up to a maximum energy. For a background magnetic field of $B sim 0.1$ G, this saturation energy is $sim 10^{16}$ eV, while for $B sim 10$ G it is $sim 10^{18}$ eV. The simulations also reveal a clear association of the accelerated particles with the regions of fast reconnection. In the early stages of the development of the non-linear growth of CDKI in the jet, when there are still no sites of fast reconnection, injected particles are also efficiently accelerated, but by magnetic curvature drift in the wiggling jet spine. However, they have to be injected with an initial energy much larger than that required for particles to accelerate in reconnection sites. Finally, we have also obtained from the simulations an acceleration time due to reconnection with a weak dependence on the particles energy $E$, $t_A propto E^{0.1}$. The energy spectrum of the accelerated particles develops a high energy tail with a power law index $p sim$ -1.2 in the beginning of the acceleration, in agreement with earlier works. Our results provide an appropriate multi-dimensional framework for exploring this process in real systems and explain their complex emission patterns, specially in the very high energy bands and the associated neutrino emission recently detected in some blazars.
We propose a thermodynamic version of the Axelrod model of social influence. In one-dimensional (1D) lattices, the thermodynamic model becomes a coupled Potts model with a bonding interaction that increases with the site matching traits. We analytica
lly calculate thermodynamic and critical properties for a 1D system and show that an order-disorder phase transition only occurs at T = 0 independent of the number of cultural traits q and features F. The 1D thermodynamic Axelrod model belongs to the same universality class of the Ising and Potts models, notwithstanding the increase of the internal dimension of the local degree of freedom and the state-dependent bonding interaction. We suggest a unifying proposal to compare exponents across different discrete 1D models. The comparison with our Hamiltonian description reveals that in the thermodynamic limit the original out-of-equilibrium 1D Axelrod model with noise behaves like an ordinary thermodynamic 1D interacting particle system.
Complete description of the singular sectors of the 1-layer Benney system (classical long wave equation) and dToda system is presented. Associated Euler-Poisson-Darboux equations E(1/2,1/2) and E(-1/2,-1/2) are the main tool in the analysis. A comple
te list of solutions of the 1-layer Benney system depending on two parameters and belonging to the singular sector is given. Relation between Euler-Poisson-Darboux equations E(a,a) with opposite sign of a is discussed.
The singular sector of zero genus case for the Hermitian random matrix model in the large N limit is analyzed. It is proved that the singular sector of the hodograph solutions for the underlying dispersionless Toda hierarchy and the singular sector o
f the 1-layer Benney (classical long wave equation) hierarchy are deeply connected. This property is due to the fact that the hodograph equations for both hierarchies describe the critical points of solutions of Euler-Poisson-Darboux equations E(a,a), with a=-1/2 for the dToda hierarchy and a=1/2 for the 1-layer Benney hierarchy.
