Category: Seminaire 2020/2021

Title: Contacts in granular materials: numerical schemes based on convex optimization problems Abstract: In this presentation we focus on the numerical simulation of granular materials: collections of macroscopic non-Brownian and rigid grains (sand, cereals, sugar, rubble…). The contacts between grains lead to singular interactions for which adapted numerical schemes must…

more

Modelling and control microalgae for strain performance improvement Microalgae are photosynthetic microorganisms whose potential has been highlighted in the last decade, especially for feed, food, and renewable energy production. When grown in photobioreactors, they can reach a higher actual photosynthetic yield compared to terrestrial plants, and thus larger biomass productivity.…

more

Title: Guided modes in thin periodic honeycomb structures. Photonic crystals are materials with a periodic structure that allows them to control the propagation of light. Due to the complexity of the structure (multiple reflections due to the periodicity of the material), destructive interference phenomena can appear at certain frequencies. More…

more

The wave localization landscape In disordered systems or systems with complex geometry, standing waves can obey a phenomenon called “localization”. This localization consists in a concentration of the wave energy in a very restricted sub-region of the whole domain, and can occur even in the absence of any classical confinement.…

more

Abstract: The results of this talk on recent numerical aspects of metric optimization have been obtained in collaboration with Chiu-Yen Kao and Braxton Osting. The first part of the talk is dedicated to Nash’s isometric embedding theorem for surfaces. We recall the impressive results obtained by HEVEA’s project for the…

more

Study of Schwarz methods in presence of crossover points Abstract Two-level domain decomposition methods were developed in the 1990s to access scalability in the presence of a large number of sub-domains. In the last decade, “coarse” spaces have been developed to deal with very heterogeneous problems. In this work, we…

more

Title : Approximation with tensor networks Tensor networks (TNs) are prominent model classes for the approximation of high-dimensional functions in computational and data science. Tree-based TNs based, also known as tree-based tensor formats, can be seen as particular feed-forward neural networks. After an introduction to approximation tools based on tree…

more

On Hamiltonian particle schemes for Vlasov-Maxwell equations. Variational discretizations are known for preserving key physical invariants in a natural way, leading to long-time stability properties. In this talk I will present a discrete action principle for the Vlasov-Maxwell equations that applies in a general structure-preserving discrete framework. In this framework…

more

Analysis and approximation of partially congested Navier-Stokes equations In this talk, I will present and analyze a mathematical model for compressible flows subject to maximum density constraint. The aim is to model saturation phenomena (congestion) corresponding to the disappearance of one of the two phases of the mixture for two-phase…

more

Depth-Adaptive Neural Networks from the Optimal Control viewpoint. Absrtact: In recent years, deep learning has been connected with optimal control problem as a way to define a notion of a continuous underlying learning problem. In this view, neural networks can be interpreted as a discretization of a parametric Ordinary Differential…

more