Leopoldo Milano

Learn More
A feedforward neural network architecture aimed at survival probability estimation is presented which generalizes the standard, usually linear, models described in literature. The network builds an approximation to the survival probability of a system at a given time, conditional on the system features. The resulting model is described in a hierarchical(More)
We study one-point statistical properties of the induced turbulent electric field for a magnetohydrodynamic (MHD) plasma under the quasinormal approximation. Assuming exact Gaussianity for both the velocity field and the magnetic field, and different degrees of correlations between their Cartesian components, we derive the probability distribution function(More)
In the last decade, the use of neural networks (NN) and of other soft computing methods has begun to spread also in the astronomical community which, due to the required accuracy of the measurements, is usually reluctant to use automatic tools to perform even the most common tasks of data reduction and data mining. The federation of heterogeneous large(More)
In this paper, a hierarchical Bayesian learning scheme for autoregressive neural network models is shown, which overcomes the problem of identifying the separate linear and nonlinear parts modeled by the network. We show how the identification can be carried out by defining suitable priors on the parameter space, which help the learning algorithms to avoid(More)
A stochastic background of gravitational waves is expected to arise from a superposition of a large number of unresolved gravitational-wave sources of astrophysical and cosmological origin. It should carry unique signatures from the earliest epochs in the evolution of the Universe, inaccessible to standard astrophysical observations. Direct measurements of(More)
In this paper, a novel information geometric-based variable selection criterion for multi-layer perceptron networks is described. It is based on projections of the Riemannian manifold defined by a multi-layer perceptron network on submanifolds defined by multi-layer perceptron networks with reduced input dimension. We show how the divergence between models(More)
(Affiliations can be found after the references in the electronic version) ABSTRACT Aims. A transient astrophysical event observed in both gravitational wave (GW) and electromagnetic (EM) channels would yield rich scientific rewards. A first program initiating EM follow-ups to possible transient GW events has been developed and exercised by the LIGO and(More)
Landslide inventories show that the statistical distribution of the area of recorded events is well described by a power law over a range of decades. To understand these distributions, we consider a cellular automaton to model a time and position dependent factor of safety. The model is able to reproduce the complex structure of landslide distribution , as(More)