Abstract of Leandro Recova's talk

We propose a regularization and variable selection method, named arbitrary rectangle-range generalized elastic net (ARGEN). It can be applied in high dimensional sparse linear regression models. We propose an algorithm to solve ARGEN; it is an extension of multiplicative updates. Multiple simulation studies and a real-world application in the stock market show that ARGEN applies to more complicated problems, outperforms and extends the lasso, ridge, and elastic net.

Nonlocal theories have emerged with powerful models and methods to analyze and predict complex phenomena. Different versions of nonlocal operators have been proposed, each with its advantages and challenges. In this talk I will give an introduction to main ideas in the nonlocality framework and present two sets of results for Helmholtz-Hodge type decompositions.

Statistical solutions are time-parameterized probability measures on spaces of integrable functions, which have been proposed recently as a framework for global solutions for multi-dimensional hyperbolic systems of conservation laws. We present a numerical algorithm to approximate statistical solutions of conservation laws and show that under the assumption of 'weak statistical scaling', which is inspired by Kolmogorov's 1941 turbulence theory, the approximations converge in an appropriate topology to statistical solutions. We will show numerical experiments which indicate that the assumption might…

In this talk, we will present a recent study of the Maxwell-Bloch equations that model the nonlinear interactions of light and matter, where the light is modeled classically by the Maxwell's equations with dispersions and the medium is modeled quantum-mechanically by the multilevel rate equations. We will show the connection between rate equations and the density matrix, where the former formulation is widely used in the engineering community and the latter in the Physics literature. A multiscale analysis of the…