HAL-Ecole des Ponts ParisTech
Not a member yet
42743 research outputs found
Sort by
Explaining gender differences in migrant sorting: Evidence from Canada-US migration
No full textThis paper uses newly digitized Canada-Vermont border crossing records from the early twentieth century to document substantial differences in how female and male migrants sorted across US destination counties by earnings potential. Income maximization largely explains sorting patterns among men. For single women, gender-based labor market constraints were important, with locations offering more work opportunities attracting women with higher earnings capacity. Among married women, destination choices were much less influenced by labor market characteristics. These findings reveal how labor market constraints based on gender and marriage influence the allocation of migrant talent across destinations
The changing face of progress: Bui Trang Chuoc's Thac Ba Hydropower Plant
No full textThis article analyzes Bui Trang Chuoc's 1975 lacquer painting "Thac Ba Hydropower Plant," exploring its artistic and historical significance. The work represents a departure from traditional Vietnamese landscape art by celebrating industrial development and Soviet-Vietnamese cooperation. The analysis gains new relevance following the dam's near-failure in September 2024, which transforms the painting from a pure celebration of progress into a necessary meditation on the relationship between technological advancement and environmental sustainability. The piece exemplifies how our understanding of development has evolved from dominance over nature toward ecological awareness.Cet article analyse la peinture laque "Centrale hydroélectrique de Thac Ba" de Bui Trang Chuoc (1975), explorant sa signification artistique et historique. L'œuvre représente une rupture avec l'art paysager vietnamien traditionnel en célébrant le développement industriel et la coopération soviéto-vietnamienne. L'analyse acquiert une nouvelle pertinence suite à la quasi-rupture du barrage en septembre 2024, transformant la peinture d'une pure célébration du progrès en une nécessaire méditation sur la relation entre avancement technologique et durabilité environnementale. L'œuvre illustre l'évolution de notre conception du développement, passant de la domination de la nature vers une conscience écologique
Machine learning assisted canonical sampling (MLACS)
No full textInternational audienceThe acceleration of material property calculations while maintaining ab initio accuracy (1 meV/atom) is one of the major challenges in computational physics. In this paper, we introduce a Python package enhancing the computation of (finite temperature) material properties at the ab initio level using machine learning interatomic potentials (MLIP). The Machine-Learning Assisted Canonical Sampling (MLACS) method, grounded in a self-consistent variational approach, iteratively trains a MLIP using an active learning strategy in order to significantly reduce the computational cost of ab initio simulations. MLACS offers a modular and user-friendly interface that seamlessly integrates Density Functional Theory (DFT) codes, MLIP potentials, and molecular dynamics packages, enabling a wide range of applications, while maintaining a near-DFT accuracy. These include sampling the canonical ensemble of a system, performing free energy calculations, transition path sampling, and geometry optimization, all by utilizing surrogate MLIP potentials, in place of ab initio calculations. This paper provides a comprehensive overview of the theoretical foundations and implementation of the MLACS method. We also demonstrate its accuracy and efficiency through various examples, showcasing the capabilities of the MLACS package
Comparison between tensor methods and neural networks in electronic structure calculations
No full textInternational audienceThis article compares the tensor method density matrix renormalization group (DMRG) with two neural network based methods -namely FermiNet and PauliNet) for determining the ground state wavefunction of the many-body electronic Schrödinger problem. We provide numerical simulations illustrating the main features of the methods and showing convergence with respect to some parameter, such as the rank for DMRG, and number of pretraining iterations for neural networks. We then compare the obtained energy with the methods for a few atoms and molecules, for some of which the exact value of the energy is known for the sake of comparison. In the last part of the article, we propose a new kind of neural network to solve the Schrödinger problem based on the training of the wavefunction on a simplex, and an explicit permutation for evaluating the wavefunction on the whole space. We provide numerical results on a toy problem for the sake of illustration.Cet article compare la méthode tensorielle du groupe de renormalisation de la matrice de densité (DMRG) avec deux méthodes basées sur les réseaux de neurones, FermiNet et PauliNet, pour déterminer la fonction d’onde de l’état fondamental du problème de Schrödinger électronique à N corps. Nous fournissons des simulations numériques illustrant les principales caractéristiques des méthodes et montrant la convergence par rapport à certains paramètres, tels que le rang pour DMRG et le nombre d’itérations de pré-entraînement pour les réseaux de neurones. Nous comparons ensuite l’énergie obtenue avec les méthodes appliquées à quelques atomes et molécules, dont nous disposons pour certains de l’énergie exacte à comparer. Dans la dernière partie de l’article, nous proposons un nouveau type de réseau de neurones pour résoudre le problème de Schrödinger basé sur l’entraînement de la fonction d’onde sur un simplexe, et une permutation explicite pour évaluer la fonction d’onde sur l’ensemble de l’espace. Nous proposons des résultats numériques sur un problème de jouet pour l’illustrer
Embedded corrector problems for homogenization in linear elasticity
No full textInternational audienceIn this article, we extend the study of embedded corrector problems, that we have previously introduced in the context of the homogenization of scalar diffusive equations, to the context of homogenized elastic properties of materials. This extension is not trivial and requires mathematical arguments specific to the elasticity case. Starting from a linear elasticity model with highly-oscillatory coefficients, we introduce several effective approximations of the homogenized tensor. These approximations are based on the solution to an embedded corrector problem, where a finite-size domain made of the linear elastic heterogeneous material is embedded in a linear elastic homogeneous infinite medium, the constant elasticity tensor of which has to be appropriately determined. The approximations we provide are proven to converge to the homogenized elasticity tensor when the size of the embedded domain tends to infinity. Some particular attention is devoted to the case of isotropic materials
Non-decreasing martingale couplings
No full textInternational audienceFor many examples of couples of probability measures on the real line in the convex order, we observe numerically that the Hobson and Neuberger martingale coupling, which maximizes for the integral of with respect to any martingale coupling between and , is still very close to maximize this integral for and minimize it for \rho>2. We investigate theoretically when it is an optimizer for and exhibit rather restrictive sufficient conditions. We also exhibit couples such that it is not an optimizer. The support of the Hobson and Neuberger coupling is known to satisfy some monotonicity property which we call non-decreasing. We check that the non-decreasing property is preserved for maximizers when . In general, there exist distinct non-decreasing martingale couplings, and we find some decomposition of which is in one-to-one correspondence with martingale couplings non-decreasing in a generalized sense
The Decline of a Great Financial Intermediary: Notaries in France, 1851–1934
No full textInternational audienceVia peer-to-peer lending, notaries mobilized immense amounts of capital in France up into the 1930s. They did so despite a thriving stock market, widespread banking with branches, and the existence of an effective lender of last resort. Using detailed evidence on their careers and their businesses, we analyze their training, their successes and failures, how they were regulated, and how they dealt with political and economic crises. What we uncover teaches broader lessons about the misconceptions surrounding modernization, financial development, and the reputational models that economists rely upon to explain informal dealings
Computation of toroidal Schnyder woods made simple and fast: from theory to practice
No full textWe consider the problem of computing Schnyder woods for graphs embedded on the torus. We design simple linear-time algorithms based on canonical orderings that compute toroidal Schnyder woods for simple toroidal triangulations. The Schnyder woods computed by one of our algorithm are crossing and satisfy an additional structural property: at least two of the mono-chromatic components of the Schnyder wood are connected. We also exhibit experimental results empirically confirming three conjectures involving the structure of toroidal and higher genus Schnyder woods.</div
Research Excellence à la Française Fifty How-Tos for Students and Professionals
No full textNational audience3rd cycl
EURAD state-of-the-art report: thermo-hydro-mechanical behaviour at high temperature of host clay formations
No full textInternational audienceMost safety cases for radioactive waste disposal concepts consider a temperature limit of 90°C in the clay host rock. Being able to tolerate higher temperature would have significant advantages. For this reason, part of the EURAD-HITEC project aimed at determining the influence of temperature above 90°C on clay host rock properties, trying to establish the possible extent of elevated temperature damage in the near and far field of clay host rock formations and the consequences of any such damage. Three clay formations considered to host radioactive waste repositories in Europe were the focus of the studies: the Boom Clay, the Callovo-Oxfordian claystone and the Opalinus Clay. A summary of the background knowledge about the thermo-hydro-mechanical behaviour of these clay host rocks is first presented. Then, the experimental and modelling activities carried out in the framework of the EURAD-HITEC project concerning these materials have been synthesised. The laboratory tests analysed the impact of temperature on the short- and long-term behaviour of the clay host rock and the self-sealing processes. Hydro-mechanical couplings between peak pore water pressure, temperature, permeability and confining stress were identified. The results confirmed that the claystone keeps its good mechanical and retention properties even when heated up to 100°C. Provided that the clay content of the samples is high enough, self-sealing was an efficient mechanism whatever the experimental conditions, although temperature may have a delaying effect. Poro-elastic models were used to model generic cases of a high-level waste repository, and consistent results were obtained by the different codes and teams, which shows the robustness of the modelling approach used to design the repositories. Two heating tests, performed in the HADES (Belgium) and MHM (France) underground research laboratories, were selected as benchmarks for the modelling activities. The evolutions of temperature and pore pressure were well modelled in the far field with a poro-elastic approach, but more advanced models are needed to take into account the processes occurring around the tunnels (e.g., modification of hydraulic properties within the EDZ, creep). The modelling of laboratory experiments showed the importance of a good understanding of the tests setup and of the boundary conditions