25353 research outputs found
Sort by
Learning from the construction of Environmental Performance Indicators in new project development: a case study
No full textInternational audienc
La gestion de l’activité de prise en charge des patients peut-elle devenir visible ?
No full textInternational audienceDespite the substantial resources invested in the healthcare system, problems related to dissatisfaction among health professionals persist. The hypothesis put forward in this article is that many issues lie at the level of real activity and the on-the-ground experiences of healthcare professionals. Drawing on an analysis of the administrative burdens, hospital hierarchies, and the coordination of patient pathways, and taking inspiration from pragmatist thinking, an overall diagnosis emerges: healthcare professionals on the ground are developing effective solutions, but these efforts often go unnoticed because they differ from the official reform initiatives in place.Malgré les ressources injectées dans le système, le malaise des professions de santé persiste. L’hypothèse avancée est qu’une grande partie des problèmes se situe au niveau de l’activité réelle. À partir d’une analyse sur le fardeau administratif, les hiérarchies hospitalières, et la coordination du parcours du patient, et en s’inspirant de la pensée pragmatiste, un diagnostic d’ensemble ressort : menée par des professionnels qui agissent sur le terrain, des solutions adaptées se bâtissent, mais restent assez invisibles, car éloignées de la mise en place des réformes officielles
Faut-il faire de l'épistémologie ?: (et doit-on préciser sa "posture épistémologique" ?)
No full textInternational audienceJ'ai l'impression de radoter à propos de cette question (Dumez, 2010 ; 2011 ; 2023b). Le malheur est qu'en 2024 (l'expérience m'a été donnée au moins en 2023), un membre de jury de thèse ou d'HDR peut encore demander au candidat de préciser quelle est sa « posture épistémologique », de lui enjoindre d'expliquer si sa recherche relève du positivisme, du constructivisme, ou de l'interprétativisme (ou du réalisme critique, etc.) et de considérer que c'est là une obligation.Est-on contraint, quand on est chercheur ou doctorant, de faire de l'épistémologie
Unravelling in Collaborative Learning
No full textInternational audienceCollaborative learning offers a promising avenue for leveraging decentralized data. However, collaboration in groups of strategic learners is not a given. In this work, we consider strategic agents who wish to train a model together but have sampling distributions of different quality. The collaboration is organized by a benevolent aggregator who gathers samples so as to maximize total welfare, but is unaware of data quality. This setting allows us to shed light on the deleterious effect of adverse selection in collaborative learning. More precisely, we demonstrate that when data quality indices are private, the coalition may undergo a phenomenon known as unravelling, wherein it shrinks up to the point that it becomes empty or solely comprised of the worst agent. We show how this issue can be addressed without making use of external transfers, by proposing a novel method inspired by probabilistic verification. This approach makes the grand coalition a Nash equilibrium with high probability despite information asymmetry, thereby breaking unravelling
Unsupervised Representation Learning for Diverse Deformable Shape Collections
No full textInternational audienceWe introduce a novel learning-based method for encoding and manipulating 3D surface meshes. Our method is specifically designed to create an interpretable embedding space for deformable shape collections. Unlike previous 3D mesh autoencoders that require meshes to be in a 1to-1 correspondence, our approach is trained on diverse meshes in an unsupervised manner. Central to our method is a spectral pooling technique that establishes a universal latent space, breaking free from traditional constraints of mesh connectivity and shape categories. The entire process consists of two stages. In the first stage, we employ the functional map paradigm to extract point-to-point (p2p) maps between a collection of shapes in an unsupervised manner. These p2p maps are then utilized to construct a common latent space, which ensures straightforward interpretation and independence from mesh connectivity and shape category. Through extensive experiments, we demonstrate that our method achieves excellent reconstructions and produces more realistic and smoother interpolations than baseline approaches. Our code can be found online: https: //github.com/Fraunhofer-SCAI/DISCO-AE
Mirroring Call-By-Need, or Values Acting Silly
No full textInternational audienceCall-by-need evaluation for the λ-calculus can be seen as merging the best of call-by-name and call-by-value, namely the wise erasing behaviour of the former and the wise duplicating behaviour of the latter. To better understand how duplication and erasure can be combined, we design a degenerated calculus, dubbed call-by-silly, that is symmetric to call-by-need in that it merges the worst of call-by-name and call-by-value, namely silly duplications by-name and silly erasures by-value. We validate the design of the call-by-silly calculus via rewriting properties and multi types. In particular, we mirror the main theorem about call-by-need - that is, its operational equivalence with call-by-name - showing that call-by-silly and call-by-value induce the same contextual equivalence. This fact shows the blindness with respect to efficiency of call-by-value contextual equivalence. We also define a call-by-silly strategy and measure its length via tight multi types. Lastly, we prove that the call-by-silly strategy computes evaluation sequences of maximal length in the calculus
Semantic Bounds and Multi Types, Revisited
No full textInternational audienceIntersection types are a standard tool in operational and semantical studies of the λ-calculus. De Carvalho showed how multi types, a quantitative variant of intersection types providing a handy presentation of the relational denotational model, allows one to extract precise bounds on the number of β-steps and the size of normal forms. In the last few years, de Carvalho’s work has been extended and adapted to a number of λ-calculi, evaluation strategies, and abstract machines. These works, however, only adapt the first part of his work, that extracts bounds from multi type derivations, while never consider the second part, which deals with extracting bounds from the multi types themselves. The reason is that this second part is more technical, and requires to reason up to type substitutions. It is however also the most interesting, because it shows that the bounding power is inherent to the relational model (which is induced by the types, without the derivations), independently of its presentation as a type system. Here we dissect and clarify the second part of de Carvalho’s work, establishing a link with principal multi types, and isolating a key property independent of type substitutions
Vertical Agrivoltaics System: What Solar Radiation to Share Between Agriculture and PV Modules?
No full textInternational audienceAgrivoltaics, which combines agricultural production with solar energy generation, addresses the water-food-energy nexus, enhancing resilience in both energy and food supply systems. This concept necessitates evaluating the synergies between agricultural and photovoltaic (PV) production. Specifically, this work delves into the physical interactions between vertically mounted bifacial PV modules and crops concerning shortwave solar radiation.The primary objective is to develop a comprehensive solar radiation model that elucidates the contributions of downwelling and upwelling solar radiation on the front and rear faces of PV modules and the ground, emphasizing dual-use applications. A major challenge lies in the spatial variability of incident and reflected solar radiation. Therefore, it's essential to consider shading geometry, the hemispherical distribution of downwelling radiance, and the temporal profile of albedo across different scales, from intra-daily to seasonal variations. Vertical PV modules are particularly sensitive to reflected radiation from crops, soil, or other PV modules, necessitating precise quantification of these components.To achieve this, we propose integrating a systematic Cell to System gain and loss methodology, adapted from Cell to Module analysis, with a matrix-based Raytracing approach (RADIANCE). This approach separates the sky's angular energy distribution from the geometry. We address the limitation of fixed surface properties in this method by binning albedo values and interpolating to capture temporal variations in ground properties. This results in an efficient transposition method compared to classic Raytracing techniques.This model describes solar radiation in the vertical plane of PV modules, accounting for incident and reflected beam and diffuse components for each face, both at minute intervals and cumulatively on a monthly scale. Validation through in-situ pyranometer measurements at minute resolution, which were quality-checked, demonstrates the model's accuracy over a period of six months. An impact transfer analysis between albedo and production is underway to discuss the temporal variability impacts
Collaborating Foundation Models for Domain Generalized Semantic Segmentation
No full texthttps://github.com/yasserben/CLOUDS ; Accepted to CVPR 2024International audienceDomain Generalized Semantic Segmentation (DGSS) deals with training a model on a labeled source domain with the aim of generalizing to unseen domains during inference. Existing DGSS methods typically effectuate robust features by means of Domain Randomization (DR). Such an approach is often limited as it can only account for style diversification and not content. In this work, we take an orthogonal approach to DGSS and propose to use an assembly of CoLlaborative FOUndation models for Domain Generalized Semantic Segmentation (CLOUDS). In detail, CLOUDS is a framework that integrates FMs of various kinds: (i) CLIP backbone for its robust feature representation, (ii) generative models to diversify the content, thereby covering various modes of the possible target distribution, and (iii) Segment Anything Model (SAM) for iteratively refining the predictions of the segmentation model. Extensive experiments show that our CLOUDS excels in adapting from synthetic to real DGSS benchmarks and under varying weather conditions, notably outperforming prior methods by 5.6% and 6.7% on averaged miou, respectively. The code is available at : https://github.com/yasserben/CLOUD
Periodic Gamma-Ray Modulation of the Blazar PG 1553+113 Confirmed by Fermi-LAT and Multiwavelength Observations
No full textInternational audienceA 2.1 yr periodic oscillation of the gamma-ray flux from the blazar PG 1553+113 has previously been tentatively identified in ∼7 yr of data from the Fermi Large Area Telescope. After 15 yr of Fermi sky-survey observations, doubling the total time range, we report >7 cycle gamma-ray modulation with an estimated significance of 4σ against stochastic red noise. Independent determinations of oscillation period and phase in the earlier and the new data are in close agreement (chance probability <0.01). Pulse timing over the full light curve is also consistent with a coherent periodicity. Multiwavelength new data from Swift X-Ray Telescope, Burst Alert Telescope, and UVOT, and from KAIT, Catalina Sky Survey, All-Sky Automated Survey for Supernovae, and Owens Valley Radio Observatory ground-based observatories as well as archival Rossi X-Ray Timing Explorer satellite-All Sky Monitor data, published optical data of Tuorla, and optical historical Harvard plates data are included in our work. Optical and radio light curves show clear correlations with the gamma-ray modulation, possibly with a nonconstant time lag for the radio flux. We interpret the gamma-ray periodicity as possibly arising from a pulsational accretion flow in a sub-parsec binary supermassive black hole system of elevated mass ratio, with orbital modulation of the supplied material and energy in the jet. Other astrophysical scenarios introduced include instabilities, disk and jet precession, rotation or nutation, and perturbations by massive stars or intermediate-mass black holes in polar orbit