Portail HAL de l'Institut Polytechnique de Paris
Not a member yet
28935 research outputs found
Sort by
Revealing the Power Dynamics of Collaborative Sense-Making Supported by Participatory Data Physicalization
No full textInternational audienceWhile it is proven that the individual construction of a data physicalization aids personal sense-making, little is known about how sense-making is negotiated when it is shared by multiple, co-located participants.Since participatory data physicalization can inadvertently prioritize dominant views, we interpreted data feminism principles to design a collaborative physicalization construction process that empowers stakeholders and participants to co-determine how meanings are represented.This process revealed how the interplay of physical and non-physical actions during construction negotiations supported collaborative sense-making among 14 groups of 55 participants during 4 workshops, enabling us to articulate how explicit power is embodied by the physicalization artifact and negotiated between authoring and collaborating participants, and facilitators; whereas tacit power operates through artifact meanings, participant identity and design decisions.By providing one operationalization of data-feminist critique into the form of design requirements, our contributions support the design of more equitable physicalization and visualization construction methods
Statistical wave field theory: Anisotropic wave fields under Neumann's boundary condition
No full textInternational audienceThe statistical wave field theory mathematically establishes the statistical laws of the solutions to the wave equation in a bounded domain. It provides the closed-form expressions of the power distribution and the correlations of the wave field jointly over time, frequency, and space, which hold at high frequency and after many reflections, in terms of the geometry and the specific admittance of the boundary surface. This theory was originally developed in the particular case of mixing rooms, which are characterized by a diffuse wave field, based on the theory of dynamical billiards and on Weyl-like asymptotic laws. Then it was extended to the finite family of special polyhedra, where the wave field is anisotropic, based on a simpler geometric approach related to mathematical crystallography. In this paper, we introduce a unified version of the theory dedicated to a class of semi-mixing billiards. In the case of Neumann's boundary condition, we show that the wave field is stationary, but it is generally anisotropic. In particular, the correlation between two spatial positions at a given frequency is different from the well-known cardinal sine formula that characterizes diffuse acoustic fields
Adding Decision Problem Makes Information More Valuable
No full textWe consider decision-making under incomplete information about an unknown state of nature. We show that a decision problem yields a higher value of information than another, uniformly across information structures, if and only if it is obtained by adding an independent, parallel decision problem.</div
Guaranteed, locally efficient, and robust a posteriori estimates for nonlinear elliptic problems in iteration-dependent norms. An orthogonal decomposition result based on iterative linearization
No full textWe consider numerical approximations of nonlinear, monotone, and Lipschitz-continuous elliptic problems, with gradient-dependent or gradient-independent diffusivity. For this purpose, we first employ a symmetric iterative linearization, with examples including respectively the Kačanov, Newton, Zarantonello and fixed-point (Picard), L, and M-schemes. We design an iteration-dependent norm, stemming from the concrete linearization and finite element discretization, which leads to an orthogonal decomposition of the total error into the linearization and discretization components. This norm is of reaction-diffusion type, so that available results from a posteriori analysis of linear (singularly perturbed) reaction-diffusion problems enable us to obtain guaranteed and locally efficient error bounds robust with respect to the strength of the nonlinearity. Moreover, the clear distinction and orthogonality of the error components leads to a design of adaptive stopping criteria for iterative linearizations ensuring a natural balance between discretization and linearization components. Under appropriate assumptions, we also show that the iteration-dependent norm is, up to a multiplicative factor tending to one, equivalent to the iteration-independent natural discrete energy norm. In an appendix, we address the Newton (nonsymmetric) linearization for the gradient-independent diffusivity case. Numerical experiments illustrate the theoretical developments
Fluid-structure Green's functions via BEM/BEM coupling for flow induced noise in arbitrary elastic geometries
No full textWe address the challenge of efficiently simulating the noise generated by the interaction of a turbulent flow noise with complex elastic structures, a coupled fluid/structure interaction (FSI) problem. Current approaches typically separate vibro-acoustic and hydro-acoustic contributions, limiting the accuracy of hydrodynamic noise predictions. To overcome this limitation, we develop a numerical method for computing a Green's function tailored to the coupled FSI problem, enabling a monolithic prediction of the radiated noise without separating the two components. This approach not only improves the accuracy of hydrodynamic noise simulations but also significantly reduces computational costs. The Green's function is constructed using a novel integral formulation and solved numerically via a coupled fast BEM/ BEM solver
Quand les implicites freinent l'insertion sociale : une réponse pédagogique pour les étudiants allophones en mobilité
No full textInternational audienceLe parcours des étudiants allophones en mobilité est jalonné de fragilités académiques, économiques et sociales. Cet article propose une étude longitudinale (sept ans) qui montre dans quelle mesure une approche culturelle, appuyée sur le numérique, aide des élèves-ingénieurs allophones (niveau Master) à surmonter les difficultés posées par la méconnaissance des implicites culturels de leur contexte d'étude. Un module de 20 heures en français langue étrangère (niveau A2-B1) a ainsi été créé (Salengros, 2010) au cours duquel les apprenants interviewent leurs pairs et explorent des éléments socio-culturels (cinéma, chanson, presse). Les analyses quantitatives et qualitatives effectuées grâce à cette recherche longitudinale mettent en évidence des résultats significatifs en termes d’apprentissage, de posture et d’attente. L’étude confirme que cette approche contribue à l'insertion sociale des étudiants allophones
A moment-based approach to the injective norm of random tensors
No full textIn this paper, we present a technically simple method to establish upper bounds on the expected injective norm of real and complex random tensors. Our approach is somewhat analogous to the moment method in random matrix theory, and is based on a deterministic upper bound on the injective norm of a tensor which might be of independent interest. Compared to previous approaches to these problems (spin-glass methods, epsilon-net techniques, Sudakov-Fernique arguments, and PAC-Bayesian proofs), our method has the benefit of being nonasymptotic, relatively elementary, and applicable to non-Gaussian models. We illustrate our approach on various models of random tensors, recovering some previously known (and conjecturally tight) bounds with simpler arguments, and presenting new bounds, some of which are provably tight. From the perspective of statistical physics, our results yield rigorous estimates on the ground-state energy of real and complex, possibly non-Gaussian, spin glass models. From the perspective of quantum information, they establish bounds on the geometric entanglement of random bosonic states and of random states with bounded multipartite Schmidt rank, both in the thermodynamic limits as well as the regimes of large local dimensions
Galois groups of low dimensional abelian varieties over finite fields
No full textInternational audienceWe consider three isogeny invariants of abelian varieties over finite fields: the Galois group, Newton polygon, and the angle rank. Motivated by work of Dupuy, Kedlaya, and Zureick-Brown, we define a new invariant called the weighted permutation representation which encompasses all three of these invariants and use it to study the subtle relationships between them. We use this permutation representation to classify the triples of invariants that occur for abelian surfaces and simple abelian threefolds
Critique écosystémique des droits humains
No full textInternational audienceL'urgence écologique exige la mise en place de contraintes sur le système productif, et donc sur les modes de consommation qui en découlent. Mais celles-ci sont bien souvent repoussées comme autant d'atteintes aux libertés fondamentales. Semble ainsi s'opérer une confusion entre liberté individuelle et mode de vie non soutenable, qui façonne les représentations individuelles, sociétales, médiatiques, politiques, jusqu'à rejaillir dans la sphère juridique où se manifestent les tensions entre protection de l'environnement et préservation des libertés fondamentales. Cet amalgame court-circuite la réflexion sur les changements sociétaux à conduire pour faire face à l'urgence écologique via le spectre de l'« écologie punitive » ou de la « dictature verte ». Le projet de recherche CritÉcoDhu -Critique écosystémique des droits humains : repenser les « droits de l'Homme » à l'aune des limites planétaires -vise à repenser la fondamentalité des droits dans le cadre d'un nouveau contrat social lié à l'urgence écologique. Ce premier ouvrage collectif, issu d'un séminaire de cadrage entre juristes, philosophes, politistes, anthropologues et climatopaléontologue, réunit tout à la fois des études de cas, des analyses inédites et des relectures critiques de textes fondamentaux qui confrontent libertés individuelles, urgence environnementale et démocratie
Double Strike: Breaking Approximation-Based Side-Channel Countermeasures for DNNs
No full textInternational audienceDeep neural networks (DNNs), which support services such as driving assistants and medical diagnoses, undergo lengthy and expensive training procedures. Therefore, the training's outcome -the DNN weights -represents a significant intellectual property asset to protect. Side-channel analysis (SCA) has recently appeared as an effective approach to recover this confidential asset from DNN implementations. In response, researchers have proposed to defend DNN implementations through classic side-channel countermeasures, at the cost of higher energy consumption, inference time, and resource utilisation. Following a different approach, Ding et al. (HOST'25) introduced MACPRUNING, a novel SCA countermeasure based on pruning, a performance-oriented Approximate Computing technique: at inference time, the implementation randomly prunes (or skips) non-important weights (i.e., with low contribution to the DNN's accuracy) of the first layer, exponentially increasing the side-channel resilience of the protected DNN implementation. However, the original security analysis of MACPRUNING did not consider a control-flow dependency intrinsic to the countermeasure design. This dependency may allow an attacker to circumvent MACPRUNING and recover the weights important to the DNN's accuracy. This paper describes a preprocessing methodology to exploit the above-mentioned control-flow dependency. Through practical experiments on a Chipwhisperer-Lite running a MACPRUNING-protected Multi-Layer Perceptron, we target the first 8 weights of each neuron and recover 96% of the important weights, demonstrating the drastic reduction in security of the protected implementation. Moreover, we show how microarchitectural leakage improves the effectiveness of our methodology, even allowing for the recovery of up to 100% of the targeted non-important weights. Lastly, by adapting our methodology, we elaborate on how the pruning mechanism, which depends on the importance of the weights, enables the circumvention of a control-flow-free MACPRUNING implementation. With this last point, we identify the pruning mechanism underlying MACPRUNING as the root of the countermeasure's vulnerability.</div