HAL Evry
Not a member yet
19237 research outputs found
Sort by
Rapport 24-12. Déterminants et impacts de la qualité sanitaire de l’alimentation sur la nutrition et la santé humaines
No full textDisponible sur Internet le 22 decembre 2024 MOTS CLÉSQualité alimentaire ; Toxicologie alimentaire ; Innocuité sanitaire des aliments ; Xénobiotiques ; Limites maximales de résidus (LMR) ; Labels de qualité alimentaire Résumé Les composés potentiellement toxiques de l'alimentation font l'objet de larges débats contradictoires pour leurs effets sur la santé. Ils regroupent les xénobiotiques provenant de la pollution et des traitements de l'agriculture, les mycotoxines et les additifs utilisés dans la transformation des aliments. Le marché agroalimentaire européen est soumis à des réglementations plus contraignantes que les marchés américains et asiatiques, avec cependant une ouverture de plus en plus grande à la mondialisation des échanges. Un groupe de travail (GT) a auditionné des experts et acteurs du domaine pour approfondir trois questions : quels sont les enseignements et les limites de la toxicologie alimentaire ? La trac ¸abilité des sources et des modes de productions est-elle une réponse à l'enjeu de sécurité alimentaire face à la mondialisation ? Les labels de qualité alimentaire sont-ils suffisants pour assurer une information objective du consommateur sur la prévention du risque ? Le GT constate une ଝ Un rapport exprime une prise de position officielle de l'Académie nationale de médecine. L'Académie dans sa séance du mardi 26 novembre 2024 a adopté le texte de ce rapport par 64 voix pour, 3 voix contre et 4 abstentions.</div
An optimal transport based embedding to quantify the distance between playing styles in collective sports
No full textInternational audienceThis study presents a quantitative framework to compare teams in collective sports with respect to their style of play. The style of play is characterized by the team's spatial distribution over a collection of frames. As a first step, we introduce an optimal transport-based embedding to map frames into Euclidean space, allowing for the efficient computation of a distance. Then, building on this frame-level analysis, we leverage quantization to establish a similarity metric between teams based on a collection of frames from their games. For illustration, we present an analysis of a collection of games from the 2021-2022 Ligue 1 season. We are able to retrieve relevant clusters of game situations and calculate the similarity matrix between teams in terms of style of play. Additionally, we demonstrate the strength of the embedding as a preprocessing tool for relevant prediction tasks. Likewise, we apply our framework to analyze the dynamics in the first half of the NBA season in 2015-2016
Intelligence artificielle dans les secteurs culturels
No full textInternational audiencePlan de l'articleQuels sont les enjeux actuels au sujet de l'intelligence artificielle (IA) dans la culture ? Qu'a prévu le règlement européen sur l'IA sur ce point ? Pourquoi cette exigence de transparence sur les sources d'entraînement ? </p
RNA-TorsionBERT: leveraging language models for RNA 3D torsion angles prediction
No full textInternational audienceMOTIVATION. Predicting the 3D structure of RNA is an ongoing challenge that has yet to be completely addressed despite continuous advancements. RNA 3D structures rely on distances between residues and base interactions but also backbone torsional angles. Knowing the torsional angles for each residue could help reconstruct its global folding, which is what we tackle in this work. This paper presents a novel approach for directly predicting RNA torsional angles from raw sequence data. Our method draws inspiration from the successful application of language models in various domains and adapts them to RNA. RESULTS. We have developed a language-based model, RNA- TorsionBERT, incorporating better sequential interactions for predicting RNA torsional and pseudo-torsional angles from the sequence only. Through extensive benchmarking, we demonstrate that our method improves the prediction of torsional angles compared to state-of-the-art methods. In addition, by using our predictive model, we have inferred a torsion angle-dependent scoring function, called TB-MCQ, that replaces the true reference angles by our model prediction. We show that it accurately evaluates the quality of near-native predicted structures, in terms of RNA backbone torsion angle values. Our work demonstrates promising results, suggesting the potential utility of language models in advancing RNA 3D structure prediction.Source code is freely available on the EvryRNA platform: https://evryrna.ibisc.univ-evry.fr/evryrna/RNA-TorsionBERT.</p
Contrôle tolérant aux défauts basé observateurs par intervalle pour systèmes LPV commutés
No full textSwitched systems have drawn considerable attention from researchers due to their ability to model a variety of practical systems. The synthesis of observers for this class of systems has gained increasing interest over the past few decades since the estimation plays a fundamental role in determining the current system states, including measured and unmeasured variables, which is crucial for fault diagnosis and control. However, the conventional observer synthesis technique may struggle to cope with the uncertainties, resulting in reduced estimation accuracy and reliability. Moreover, in practical engineering applications, it is inevitable to encounter unknown inputs and faults due to unpredictable external disturbances, measurement noise, and potential actuator faults. Such factors may cause system performance degradation, instability, or even catastrophic failures. It is, therefore essential to enhance the system safety and reliability by developing well-designed algorithms that can effectively estimate and compensate for faults affecting the performance. The objective of this research is to provide some contributions to the state-of-the-art in the field of robust state estimation and fault-tolerant control (FTC) for a class of switched systems by addressing the aforementioned problems. The present research mainly focuses on addressing the challenge of robust state bounding estimation using interval observer techniques for uncertain switched systems subject to unknown but bounded exogenous disturbances and/or measurement noise. Based on the cooperativity system theory, the methodology presents novel interval observer structures that provide notable improvement, particularly in enhancing the robustness and accuracy of state estimation under uncertain conditions. In contrast to conventional observers may struggle to deal with uncertainties, the proposed observer can effectively cope with the problems by offering guaranteed lower and upper bounds of state estimations. In addition to robust state estimation, the thesis also focuses on the synthesis of active fault-tolerant control (AFTC) strategies designed to preserve system stability and ensure desired performance levels, even in the presence of faults. The approach employs a co-design methodology, which integrates the design of observers and controllers into a cohesive framework. This integrated design approach considers the bi-directional interaction between the estimation process and control actions, leading to the optimized overall system performance and enhanced resilience to faults.Sufficient conditions for proving the existence of the interval observers and controllers are formulated in terms of Linear Matrix Inequalities (LMIs) constraints. These conditions are derived through a combination of Lyapunov theory and Input-to-State-Stability (ISS) under the Average-Dwell Time (ADT) concept. Finally, to validate the efficacy of the proposed interval observer structures and the synthesized control laws, an application to the vehicle lateral dynamics model is presented, using MATLAB Simulink. The simulations validate the robustness of the interval observer structures and fault-tolerant control strategies, showing that the proposed approach can effectively maintain vehicle stability and control even in challenging and unpredictable environments. The results highlight the ability of the interval observers to accurately bound state estimates despite uncertainties, while the synthesized control laws successfully ensure system stability and performance even in the occurrence of faults.Les systèmes à commutation attirent de plus en plus l'attention dans le domaine scientifique grâce à leur capacité à représenter des phénomènes complexes issus de diverses applications en ingénierie. L'estimation d'état joue un rôle clé dans ces systèmes, notamment pour le diagnostic de pannes et le contrôle. Toutefois, les méthodes classiques d'observateurs peinent à traiter efficacement les incertitudes, limitant ainsi la précision et la fiabilité des résultats. Dans les applications industrielles, les perturbations externes, le bruit de mesure ou les défaillances d'actionneurs engendrent fréquemment des entrées inconnues ou des défauts. Ces aléas peuvent dégrader les performances, provoquer une instabilité, voire conduire à des dysfonctionnements critiques. Il devient donc primordial de développer des solutions algorithmiques robustes pour estimer et compenser ces anomalies, améliorant ainsi la sûreté de fonctionnement. L'un des axes principaux de cette étude concerne l'estimation d'état robuste par intervalles, une approche novatrice permettant de borner les états malgré des perturbations bornées ou un bruit de mesure. Fondée sur la théorie des systèmes coopératifs, la méthodologie proposée garantit des bornes inférieures et supérieures aux estimations, offrant une robustesse accrue face aux incertitudes par rapport aux observateurs conventionnels. Parallèlement, des stratégies de commande active tolérante aux défauts (AFTC) sont développées pour maintenir la stabilité et les performances du système, même en présence de défauts. Une approche de co-conception intègre observateurs et contrôleurs dans un cadre unifié, tenant compte des interactions entre estimation et commande. Cette synergie optimise la résilience globale et les performances. Les conditions d'existence des observateurs par intervalles et des contrôleurs sont exprimées via des inégalités matricielles linéaires (LMIs), dérivées de la théorie de Lyapunov et de la stabilité entrée-état (ISS) sous contraintes de temps de commutation moyen (ADT). Enfin, les applications à la dynamique latérale d'un véhicule, simulée sous MATLAB/Simulink, valide l'efficacité des solutions proposées. Les résultats démontrent que les observateurs par intervalles maintiennent des estimations précises malgré les incertitudes, tandis que les lois de commande AFTC préservent la stabilité et les performances, même dans des scénarios dégradés. Ces contributions renforcent la fiabilité des systèmes complexes soumis à des environnements imprévisibles
Glucagon and insulin production in pancreatic cells modeled using Petri nets and Boolean networks.
No full textIntroduced during PNSE’25, International Workshop on Petri Nets and Software Engineering, 2025Diabetes is a civilization chronic disease characterized by a constant elevated concentration of glucose in the blood. Many processes are involved in the glucose regulation, and their interactions are very complex. To better understand those processes we set ourselves a goal to create a Petri net model of the glucose regulation in the whole body. So far we have managed to create a model of glycolysis and synthesis of glucose in the liver, and the general overview models of the glucose regulation in a healthy and diabetic person.In this paper we introduce Petri nets models of insulin secretion in β cell of the pancreas, and glucagon in the pancreas α cells. Those two hormones have mutually opposite effects: insulin preventing hyperglycemia, and glucagon preventing hypoglycemia. Understanding the mechanisms of insulin and glucagon secretion constitutes the basis for understanding diabetes. We also present a model in which both processes occur together, depending on the blood glucose level. The dynamics of each model is analysed. Additionally, we transform the overall insulin and glucagon secretion system to a Boolean network, following standard transformation rules
Leaky-Integrator Echo State Network Incremental ISS Stability Analysis
No full textInternational audienceThis paper proposes a novel incremental input-to-state stability condition for a discrete-time leaky-integrator echo state network. The derived condition is further utilized for control design through Linear Matrix Inequalities (LMIs). The corresponding observer design LMI condition is also derived. A numerical simulation showcases the effectiveness of the proposed approach
Kevin Mellet, Sociologie du Marketing, (La découverte, 2023)
No full textInternational audienceRecension de "Kevin Mellet, Sociologie du Marketing, (La découverte, 2023)
Trade-offs in automating platform regulation by algorithm: evidence from a health emergency
No full textInternational audienceDigital platforms have experienced pressure to restrict and regulate sensitive ad content. In a static environment, algorithms can help platforms more quickly and easily achieve regulatory compliance. However, in dynamic contexts, the performance of algorithmic decision-making for regulatory compliance is less understood. We aim to fill this gap by exploring how algorithmic rules governing digital platforms respond to rapid environmental changes, specifically in the context of a global health crisis. We study the effect of algorithmic regulation of ad content in times of rapid change where digital ad venues need to identify sensitive ads that should be subject to more restrictive policies and practices. Our results show that ads run by governmental organizations designed to inform the public about COVID-19 are more likely to be banned by Meta's algorithm than similar ads run by non-governmental organizations. Using a difference-indifferences (DiD) approach by exploiting an algorithmic incident on Meta in March 2020, we provide evidence of platform-level mechanisms at play. After the incident, we find that the proportion of disqualified ads decreased significantly. Further analysis reveal that (mis)classification of ads is responsible for this high proportion of disqualified ads, ruling out advertiser-effects and suggesting algorithmic (mis)classification. Using human-based classification, we show that the algorithm is likely to misclassified 12% of ads related to issues of national significance. This finding challenges the notion that algorithmic decision-making is always efficient or unbiased, especially in dynamic circumstances. Overall, our study contributes to the broader conversation about algorithmic decision-making in management. We suggest that algorithmic inflexibility towards categorization in periods of unpredictable shifts worsens the problems of trying to achieve regulatory compliance using algorithms