Portail HAL-PSL
Not a member yet
293219 research outputs found
Sort by
Embedding Learning on Multiplex Networks for Link Prediction
No full textOver the past years, embedding learning on networks has shown tremendous results in link prediction tasks for complex systems, with a wide range of real-life applications. Learning a representation for each node in a knowledge graph allows us to capture topological and semantic information, which can be processed in downstream analyses later. In the link prediction task, high-dimensional network information is encoded into low-dimensional vectors, which are then fed to a predictor to infer new connections between nodes in the network. As the network complexity (that is, the numbers of connections and types of interactions) grows, embedding learning turns out increasingly challenging. This review covers published models on embedding learning on multiplex networks for link prediction. First, we propose refined taxonomies to classify and compare models, depending on the type of embeddings and embedding techniques. Second, we review and address the problem of reproducible and fair evaluation of embedding learning on multiplex networks for the link prediction task. Finally, we tackle evaluation on directed multiplex networks by proposing a novel and fair testing procedure. This review constitutes a crucial step towards the development of more performant and tractable embedding learning approaches for multiplex networks and their fair evaluation for the link prediction task. We also suggest guidelines on the evaluation of models, and provide an informed perspective on the challenges and tools currently available to address downstream analyses applied to multiplex networks
Euclid. Properties and performance of the NISP signal estimator
No full textInternational audienceThe Euclid spacecraft, located at the second Lagrangian point of the Sun-Earth system, hosts the Near-Infrared Spectrometer and Photometer (NISP) instrument. NISP is equipped with a mosaic of 16 HgCdTe-based detectors to acquire near-infrared photometric and spectroscopic data. To meet the spacecraft's constraints on computational resources and telemetry bandwidth, the near-infrared signal is processed onboard via a dedicated hardware-software architecture designed to fulfil the stringent Euclid's data-quality requirements. A custom application software, running on the two NISP data processing units, implements the NISP signal estimator: an ad-hoc algorithm which delivers accurate flux measurements and simultaneously estimates the quality of signal estimation through the quality factor parameter. This paper investigates the properties of the NISP signal estimator by evaluating its performance during the early flight operations of the NISP instrument. First, we revisit the assumptions on which the inference of the near-infrared signal is based and investigate the origin of the main systematics of the signal estimator through Monte Carlo simulations. Then, we test the flight performance of the NISP signal estimator. Results indicate a systematic bias lower than 0.01 e/s for 99% of the NISP pixel array, well within the noise budget of the estimated signal. We also derive an analytical expression for the variance of the NISP signal estimator, demonstrating its validity, particularly when the covariance matrix is not pre-computed. Finally, we provide a robust statistical framework to interpret the QF parameter, analyse its dependence on the signal estimator bias, and show its sensitivity to cosmic ray hits on NISP detectors. Our findings corroborate previous results on the NISP signal estimator and suggest a leading-order correction based on the agreement between flight data and simulations
Face à l’IA, les médias ne pourront pas faire l’économie d’une profonde remise en question de leur mode de fonctionnement
No full textL’intelligence artificielle remet profondément en cause la raison d’être des médias et leur modèle d’affaires. Leur survie pourrait passer par une rapide réflexion sur ce qui fait leur cœur de métier, à l’image de la politique suivie par le New York Times. Au risque de sombrer dans des offres très low cost ou de devenir des fournisseurs des géants de l’IA
Scholies et lexiques techniques: Actes des deux journées d’étude du même nom, tenues à Sorbonne Université les 20 mai 2022 et 27 mars 2023
No full textInternational audienc
Genomic differentiation of the black-chinned tilapia (Sarotherodon melanotheron heudelotii Duméril, 1859) along a fresh-to-hypersaline water gradient
No full textInternational audienceThe black-chinned tilapia (Sarotherodon melanotheron) is an African fish species, found in freshwater, brackish, marine and especially hypersaline (up to 110‰) habitats in Senegambia. Using 16,786 filtered single nucleotide polymorphism (SNP) markers, we investigated whether it has responded adaptively to this fresh-to-hypersaline water gradient. Significant genetic differentiation between samples was observed, revealing an interplay between geographic and environmental variation. We focused on a set of 255 outlier SNPs indicative of adaptive variation, 119 of which mapped to annotated genes in the Oreochromis niloticus genome. Significant enrichment was found for physiological pathways relevant to osmosensing and osmoregulation (e.g., inositol phosphate, thyroid hormone synthesis pathways), but also for immune-related pathways that could be activated by ion fluxes (e.g., inflammasome). Some outlier loci mapped to genes that are known to respond to salinity variation in other organisms, including genes found to be differentially expressed in black-chinned tilapia. Adaptive variation along a fresh-to-hypersaline water gradient is well supported in black-chinned tilapia, but its association with climate change specifically induced by hypersalinity deserves further attention particularly in the context of increasing cases of inverted estuaries being reported worldwide
Royautés
No full textInternational audienceÀ travers une large gamme de sources originales datant du Moyen Âge et de l’époque moderne (textes, monuments, monnaies, peintures et autres artefacts), ce livre présente une sélection d’expériences royales ancrées dans des territoires très variés allant de la Suède à l’Asie du Sud, en passant par la France, la péninsule Ibérique, le Maghreb, la Valachie et la Moldavie, ou encore l’Éthiopie. Il les rapproche et les compare en huit thématiques originales qui permettent d’aborder les royautés de façon neuve, du nom du roi à la symbolique animale, et qui visent à comprendre la pratique du pouvoir dans les sociétés du passé. Sa rédaction est le fruit d’une aventure intellectuelle inédite durant laquelle huit spécialistes ont mêlé leurs connaissances pour offrir au lecteur des documents, des commentaires et des analyses qui rendent compte du fonctionnement de ces monarchies entre le XIe et le XIXe siècle. Avec ses courts chapitres qui, tous, s’appuient sur des matériaux propres à chaque terrain, ce livre polyphonique propose une nouvelle manière d’appréhender l’histoire de la royauté
A new open boundary condition for Boussinesq-type models, applied to irregular wave fields
No full textInternational audienceWe present a novel approach to handle open boundary conditions for a Boussinesq-type wave model coupled with the nonlinear shallow water equations. Traditional methods for managing open boundaries — such as sponge layers and source functions — are computationally intensive and require ad hoc calibration. To address this, we reformulate the Boussinesq equations as a system of conservation laws with nonlocal flux and a rapidly decaying source term. This reformulation is adapted to generate waves at the boundary of the numerical domain, from surface elevation data in situations where both incoming and outgoing waves are present. The proposed numerical scheme employs a MacCormack prediction-correction strategy combined with finite volume and finite difference methods, preserving key physical properties and ensuring stability. Comparison with laboratory experiments demonstrates that our approach avoids boundary reflection issues. In particular, it is able to accurately reproduce infragravity waves associated with a random wave field propagating over a sloping beach. This work opens important perspectives for improving phase-resolving coastal wave models, with the aim of forecasting complex random wave conditions in natural environments
Infrared nanocrystals for space application: hardness to irradiations
No full textInternational audienceColloidal nanomaterials are now a cost-effective strategy for the design of infrared imagers. Their increased maturity brings them to the point where the evaluation of their potential for space and astronomy applications becomes relevant. The first step is to test their robustness against irradiation, since this process is responsible for the performance degradation of most imagers. In this study, we explore the potential of HgTe nanocrystals for short-wave infrared sensing under Xray and He⁺ ion irradiation. To first assess the degradation of the material itself, we test the evolution of a single-pixel photoconductive device upon irradiation and observe that only marginal degradation occurs for doses relevant to space applications, and degradation only occurs at a very high threshold, making this colloidal semiconductor quite promising to operate in harsh radiation environments. Complementary spectroscopic studies reveal very different degradation mechanisms upon X-ray (leading to oxidation) and He⁺ ion (sintering particles). We tested the potential of this material in a focal plane array and observed degradation, though not preventing imaging, in a range of doses where no effect was observed for the simple photoconductive device. This suggests that the CMOS readout circuit might be the bottleneck regarding operation under radiation
