HAL-Polytechnique
Not a member yet
51406 research outputs found
Sort by
Near-extremal holographic charge correlators
No full textInternational audienceWe analytically compute the low-temperature charge correlators in near-extremal black holes with a planar horizon and an infrared extremal geometry, finding excellent agreement with numerical calculations. The analytical result consistently describes the crossover between the hydrodynamic diffusive regime at low frequencies and wavenumbers, and the quantum, zero-temperature regime at high frequencies and wavenumbers. We analytically resolve the successive collisions between the diffusive pole and the non-hydrodynamic poles sourced by the infrared geometry. We demonstrate that in the limit, a pair of gapless poles survive with a dispersion relation . The nonanalytic contributions arise from the interplay with the branch cut formed by the condensation of the non-hydrodynamic poles. The real part is caused by the `snatching' of one of the non-hydrodynamic poles by the hydrodynamic diffusive pole
Tailgater, Mouthpiece, or Magnifying glass? Italian Political Journalism in the Face of Populist Radical Right News Politicization
No full textInternational audienceThe rise of populist radical right (PRR) parties has prompted extensive research on the news media's role in mainstreaming their discourse. Yet, less attention has been devoted to the emergence of diversified approaches to PRR coverage across different professional segments of national journalistic landscapes. This paper analyzes how Italian newsrooms covered the politicization of a rape initiated by PRR leaders during the 2022 election campaign. Through a mixed-methods analysis of news articles, I show how outlets' approaches to the PRR's 'talk scandal' and the ensuing controversy diverge along two key dimensions: the space granted to politicians' discourse through quotes, and the prevailing news frames adopted, either echoing or marking a distance from politicians' narratives. Differences emerge at the intersection of outlets' ideological leaning and legacy vs. new-generation status. I identify three ideal-typical tendencies: a 'tailgater' approach in legacy, left-leaning newspapers; a 'mouthpiece' model among right-wing outlets, legacy and new-generation; and a 'magnifying glass' approach in new-generation and progressive newsrooms. This highlights how within-country variations in journalists' standards for treating political discourse shape dynamics of media mainstreaming of PRR narratives
modélisation statistique pour l'aide à la décision en sports collectifs
No full textIn this thesis, we develop statistical frameworks for sports analytics that integrate temporal, spatial, and financial dimensions. The aim is to enhance performance evaluation and strategic decision-making in collective sports, with a focus on the case of football. Our work is organized around four research questions.First, we propose a novel modeling framework based on multivariate Hawkes processes to assess how individual players contribute to the generation of offensive threats. By leveraging the immigration--birth interpretation of Hawkes processes, we introduce new metrics that decompose a player’s influence into direct and indirect contributions, thereby revealing the hierarchical nature of in-game events.Second, we address the representation of spatial information in collective sports by introducing an optimal transport-based embedding of player configurations. By mapping each frame—viewed as a discrete probability measure—into a Euclidean space via successive projections, our method yields a permutation-invariant representation that preserves an interpretable notion of distance through the sliced-Wasserstein metric. This embedding facilitates many learning tasks. In particular, it enables us to measure the degree of similarity between the playing styles of two teams.Third, we develop a probabilistic model to systematically estimate team formations and quantify the rate of role swaps. Modeling observed player positions as a random permutation of latent role positions, our approach enables the identification of a clear team structure. The challenge of the cardinality of the set of permutations is addressed using a sparse selection procedure based on an overlap criterion.Fourth, we combine performance statistics with player-specific attributes using regression techniques, such as Lasso and Random Forests, to predict future market values. This analysis identifies key predictors that drive player valuations, offering actionable insights for talent evaluation and transfer strategies.Overall, the contributions of this thesis advance the state-of-the-art in sports analytics by providing frameworks that capture the intricate dynamics of in-game events, spatial configurations, and economic valuation in modern football. The proposed methodologies are validated on extensive datasets from professional leagues, demonstrating both their practical utility and theoretical robustness.Dans cette thèse, nous développons des méthodes statistiques pour l'analyse de données dans les sports collectifs. Nous nous intéressons à la fois aux dynamiques spatio-temporelles des joueurs et à l'aspect financier du jeu. Notre but est de fournir des outils pour améliorer la prise de décision dans le sport, en portant une attention particulière au cas du football. Nos travaux s'articulent autour de quatre questions de recherche.Dans le premier chapitre, nous proposons une nouvelle modélisation des événements ayant lieu dans un match de football, basée sur des processus de Hawkes. Cela a pour objectif d'évaluer la contribution individuelle à la génération de menaces offensives. Grâce à l'interprétation immigration--naissance des processus de Hawkes, nous introduisons des métriques qui décomposent l'influence d'un joueur en contributions directes et indirectes, révélant ainsi la nature hiérarchique des événements au cours d'un match.Dans le deuxième chapitre, nous abordons la question de la modélisation de l'information spatiale sur le positionnement des joueurs à un instant donné. Pour cela, nous proposons une représentation des configurations de joueurs à chaque instant basée sur le transport optimal. Nous utilisons des projections successives afin d'obtenir une représentation dans un espace euclidien qui préserve une notion de distance interprétable grâce à la métrique sliced-Wasserstein. Ce procédé facilite de nombreuses tâches d'apprentissage. En particulier, il nous permet de mesurer le degré de similarité entre les styles de jeu d'équipes différentes.Dans le troisième chapitre, nous développons un modèle probabiliste visant à identifier systématiquement les formations caractéristiques de chaque équipe et à quantifier le taux de changement de rôles entre les joueurs. En modélisant les positions observées comme une permutation aléatoire de positions latentes associées à des rôles, notre approche permet de retrouver une structure spatiale claire et intelligible. Notre méthodologie incorpore la possibilité de changements de régime, permettant ainsi de segmenter les matchs en différentes phases de jeu selon la structure de l'équipe. Le problème de la cardinalité de l'ensemble des permutations est réglé au moyen d'une procédure de sélection parsimonieuse fondée sur un critère de recouvrement.Dans la dernière partie, nous combinons des indicateurs de performance avec des attributs spécifiques aux joueurs afin de prédire les valeurs marchandes futures. Pour cela, nous utilisons des méthodes de régression telles que le Lasso et les forêts aléatoires. Cette analyse met en évidence les principaux prédicteurs influençant la valorisation des joueurs, fournissant ainsi des indications concrètes pour l'évaluation des talents et l'élaboration de stratégies de transfert
The Search-Coil Magnetometer (SCM) of the Radio and Plasma Waves Investigation (RPWI) onboard the ESA JUICE mission: in-flight performance and first observations.
No full textInternational audienceThe JUpiter ICy moons Explorer (JUICE) mission is the first large-class (L1) mission of ESA Cosmic Vision. JUICE has been launched in April 2023 with an arrival at Jupiter in 2031 and at least four years making detailed plasma observations of Jupiter's magnetosphere and of three of its largest moons (Ganymede, Callisto and Europa). The Radio and Plasma Wave Investigation (RPWI) consortium will carry the most advanced set of electric and magnetic fields sensors ever flown in Jupiter's magnetosphere, which will allow to characterize the radio emission and plasma wave environment of Jupiter and its icy moons. The Search Coil Magnetometer (SCM) of RPWI, combined with the RPWI Low-Frequency receiver (LF), will provide for the first time three-dimensional measurements of magnetic field fluctuations within Jupiter's magnetosphere, with high sensitivity (~10 fT / √Hz at 1 kHz) in the frequency range 0.1 Hz - 20 kHz. Here we present SCM in-flight performance and first observations obtained during its cruise phase, including those from the Lunar-Earth Gravity Assist (LEGA) in 2024. These observations show a nominal functioning and performance of SCM, in agreement with ground calibrations, together with a rather good magnetic cleanliness of the JUICE spacecraft. Observations during LEGA have also allowed to properly identify a number of plasma boundaries in the Earth’s magnetosphere, such as the magnetopause and the magnetotail current sheet, successfully testing the SCM capability to study such boundaries at Jupiter’s and of Ganymede's magnetosphere
Reconstruire une filière par les savoir-faire dans un secteur traditionnel : le cas de la marque de jeans 1083
No full textInternational audienc
Saisir la polyvocalité des archives audiovisuelles et numériques en contexte post-colonial
No full textCe document est le bilan du travail de l'équipe française dans le cadre du programme européen PICCH. L'objectif global du PICCH est d'engendrer une polyvocalité qui peut être incorporée dans l'archive elle-même, offrant de nouvelles façons d'entrer et d'explorer le passé par le biais d'un cadre interprétatif contemporain. À cet effet, des technologies de pointe seront utilisées pour étudier comment faire le lien entre les langues archivistiques et contemporaines, et pour soutenir l'exploration transnationale de plusieurs archives via une seule interface utilisateur interactive
Supersymétrie linéaire et non linéaire en théorie des champs et des cordes
No full textThe current state of high-energy physics is characterized by a division into two distinct branches. On one side, we have the Standard Model of particle physics, which has been tested with great precision and confirmed as the theory describing the physics at the TeV scale. On the other side, it is clear that this theory is incomplete. First, it must be coupled to gravity, following the principles of Einstein's general relativity. Second, a quantum description of gravitational interactions must be developed, just as with all other forces. The only example that has consistently realized this unification so far is String Theory, which lives near the Planck scale. Following the effective field theory paradigm, one expects these two branches - string theory in the UV and the Standard Model in the IR of the energy spectrum - to be linked through a series of lower-energy effective models, each associated with a specific energy scale between the TeV and the Planck mass, gradually revealing the features of the quantum theory of gravity. However, how to explicitly realize this connection is far from clear and well-understood.Bridging this gap is one of the main challenges of modern days high-energy physics. In this respect, a crucial role is played by supersymmetry. The reason is that supersymmetry has strong motivations to be realized on both branches of the energy spectrum we seek to connect. On the field theory side, it represents the only possible extension of the Poincare group, and its realization as a local symmetry automatically yields the gravitational interaction. On the string side, it is an essential element of the worldsheet theory in order to describe fermionic excitations and to ensure the stability of string vacua, and, moreover, the field theory limit of such superstring theories are supergravity theories. In addition, supersymmetry provides also with a well-defined framework in which to embed the effective flow from String Theory to the Standard Model. Being the latter obviously non-supersymmetric, supersymmetry must be spontaneously broken at some stage, and the scale of supersymmetry breaking becomes the vehicle with which the UV physics propagates at lower energies.In this thesis we investigate effective field and string theories, in which supersymmetry is realized - and broken - in different ways. The first construction of this type that we present are effective theories with nonlinearly realized supersymmetry, obtained through the formalism of constrained superfields. We prove a criterion to identify inconsistent constraints, we formulate, their improved version, and propose, along these lines, improved supergravity models of inflation. Then, we discuss a peculiar phenomenological application of such a framework, the gravitational production of massive gravitinos in time-dependent backgrounds, highlighting some physical subtleties of its standard description, particularly severe in the case of the gravitino.The focus moves then to pure supergravity in four dimensions. In this context, we analyze the leading-order coupling to supergravity of the rigid multiplet of the massive spin-2 field, by means of the supercurrent superfield formalism. The result of our analysis is that there exist a single consistent class of such couplings. This is fundamentally distinct from the one obtained by compactifying higher-dimensional supergravity, and it is interestingly associated to the highest-possible cutoff scale that these effective theories can reach.Finally, we consider type II B string theory and construct a novel orientifold projection, of the Scherk-Schwarz type, which breaks supersymmetry entirely. The salient features of this theory are that the O-planes couple only to the twisted sector and that it is invariant under S-duality, which can be interpret as the result of an F-theory compactification. We also construct the D-branes spectrum, analyze their stability, and compare with previous similar constructions.L'état actuel de la physique des hautes énergies se divise en deux branches. D'un côté, nous avons le Modèle Standard des particules, testé avec grande précision et confirmé comme théorie décrivant la physique à l'échelle du TeV. De l'autre, cette théorie est incomplète. D'abord, elle doit être couplée à la gravité, selon la relativité générale d'Einstein. Ensuite, une description quantique des interactions gravitationnelles doit être développée, comme pour les autres forces. Le seul exemple qui a unifié cela jusqu'ici est la théorie des cordes, proche de l'échelle de Planck. Selon le paradigme de la théorie effective des champs, on attend que ces deux branches — la théorie des cordes dans l'UV et le Modèle Standard dans l'IR du spectre énergétique — soient liées par une série de modèles effectifs à basse énergie, chacun ayant une échelle d'énergie entre le TeV et la masse de Planck, révélant progressivement les traits de la théorie quantique de la gravité. Cependant, comment établir explicitement cette connexion reste flou.Combler cet écart est l'un des grands défis de la physique moderne. La supersymétrie y joue un rôle crucial, car elle est motivée pour être réalisée sur les deux branches du spectre énergétique à connecter. Du côté des champs, elle est la seule extension possible du groupe de Poincaré, et sa réalisation en tant que symétrie locale entraîne automatiquement l'interaction gravitationnelle. Du côté des cordes, elle est clé pour la théorie de la feuille monde, afin de décrire les excitations fermioniques et garantir la stabilité des vacuums des cordes. En outre, la limite en théorie des champs des supercordes est une théorie de supergravité. La supersymétrie fournit aussi un cadre pour relier la théorie des cordes au Modèle Standard. Ce dernier, non supersymétrique, nécessite que la supersymétrie soit brisée spontanément, et l’échelle de cette rupture devient un vecteur pour la propagation de la physique UV à plus faibles énergies.Dans cette thèse, nous étudions des théories effectives des champs et des cordes, où la supersymétrie est réalisée — et brisée — de différentes manières. La première construction que nous présentons est celle des théories avec une supersymétrie réalisée de manière non linéaire, obtenue via le formalisme des superchamps contraints. Nous prouvons un critère pour identifier des contraintes inconsistantes, proposons une version améliorée de celles-ci, et suggérons, dans ce cadre, des modèles améliorés de supergravité pour l'inflation. Ensuite, nous discutons d’une application particulière, la production gravitationnelle de gravitinos massifs dans des arrière-plans dépendant du temps, en soulignant des subtilités physiques de sa description standard, surtout dans le cas du gravitino.L’attention se tourne ensuite vers la supergravité pure en quatre dimensions. Nous analysons le couplage au premier ordre de la supergravité du multiplet rigide du champ spin-2 massif, par le formalisme du superchamp de supercourant. Notre analyse montre qu’il existe une seule classe cohérente de tels couplages, fondamentalement différente de celle obtenue par compactification de la supergravité en dimensions supérieures, et intéressante car elle est liée à la plus haute échelle de coupure que ces théories effectives peuvent atteindre.Enfin, nous considérons la théorie des cordes de type II B et construisons une nouvelle projection orientifold, de type Scherk-Schwarz, qui brise complètement la supersymétrie. Les caractéristiques principales de cette théorie sont que les O-plans se couplent uniquement au secteur tordu et qu’elle est invariante sous la dualité S, ce qui peut être vu comme le résultat d’une compactification de la théorie F. Nous construisons aussi le spectre des D-branes, analysons leur stabilité et les comparons avec des constructions similaires passées
Asymptotic Analysis of a bi-monomeric nonlinear Becker-Döring system
No full textInternational audienceTo provide a mechanistic explanation of sustained then damped oscillations observed in a depolymerisation experiment, a bi-monomeric variant of the seminal Becker-D\"oring system has been proposed in~(Doumic, Fellner, Mezache, Rezaei, J. of Theor. Biol., 2019). When all reaction rates are constant, the equations are the following:\begin{align*}\frac{dv}{dt} & =-vw+v\sum_{j=2}^{\infty}c_{j}, \qquad\frac{dw}{dt} =vw-w\sum_{j=1}^{\infty}c_{j}, \\\frac{dc_{j}}{dt} & =J_{j-1}-J_{j}\ \ ,\ \ j\geq1\ \ ,\ \ \J_{j}=wc_{j}-vc_{j+1}\ \ ,\ \ j\geq1\ \ ,\ J_{0}=0,\end{align*}where and are two distinct unit species, and represents the concentration of clusters containing units. We study in detail the mechanisms leading to such oscillations and characterise the different phases of the dynamics, from the initial high-amplitude oscillations to the progressive damping leading to the convergence towards the unique positive stationary solution. We give quantitative approximations for the main quantities of interest: period of the oscillations, size of the damping (corresponding to a loss of energy), number of oscillations characterising each phase. We illustrate these results by numerical simulation, in line with the theoretical results, and provide numerical methods to solve the system
Timelike Compton scattering on a spin-0 target with kinematic twist-4 precision
No full textInternational audienceWe calculate the kinematic twist-3 and twist-4 corrections to the leading-order amplitude of timelike Compton scattering (TCS) on a (pseudo)scalar target, in the recently developed framework based on the conformal operator-product expansion. This allows us to compute the complete set of helicity amplitudes of the process, in particular those that vanish at leading twist. We compare the effects of higher-twist contributions to TCS with those in deeply virtual Compton scattering. Our estimates, based on a -meson GPD model, indicate that these contributions are sizeable and will play a crucial role in the interpretation of data from current and forthcoming experiments
Renal Transplant Survival Prediction From Unsupervised Deep Learning-Based Radiomics on Early Dynamic Contrast-Enhanced MRI
No full textInternational audienceRationale and ObjectivesEnd-stage renal disease is characterized by an irreversible decline in kidney function. Despite a risk of chronic dysfunction of the transplanted kidney, renal transplantation is considered the most effective solution among available treatment options. Clinical attributes of graft survival prediction, such as allocation variables or results of pathological examinations, have been widely studied. Nevertheless, medical imaging is clinically used only to assess current transplant status. This study investigated the use of unsupervised deep learning-based algorithms to identify rich radiomic features that may be linked to graft survival from early dynamic contrast-enhanced magnetic resonance imaging data of renal transplants.Materials and MethodsA retrospective cohort of 108 transplanted patients (mean age 50 +/- 15, 67 men) undergoing systematic magnetic resonance imaging follow-up examinations (2013 to 2015) was used to train deep convolutional neural network models based on an unsupervised contrastive learning approach. 5-year graft survival analysis was performed from the obtained artificial intelligence radiomics features using penalized Cox models and Kaplan–Meier estimates.ResultsUsing a validation set of 48 patients (mean age 54 +/- 13, 30 men) having 1-month post-transplantation magnetic resonance imaging examinations, the proposed approach demonstrated promising 5-year graft survival capability with a 72.7% concordance index from the artificial intelligence radiomics features. Unsupervised clustering of these radiomics features enabled statistically significant stratification of patients (p=0.029).ConclusionThis proof-of-concept study exposed the promising capability of artificial intelligence algorithms to extract relevant radiomics features that enable renal transplant survival prediction. Further studies are needed to demonstrate the robustness of this technique, and to identify appropriate procedures for integration of such an approach into multimodal and clinical settings