HAL-UNICE
Not a member yet
30584 research outputs found
Sort by
Low-rank tensor recovery using sequentially optimal modal projections in iterative hard thresholding (SEMPIHT)
No full text© 2017, Society for Industrial and Applied MathematicsInternational audienceIterative hard thresholding (IHT) is a simple and effective approach to parsimonious data recovery. Its multilinear rank (mrank)-based application to low-rank tensor recovery (LRTR) is especially valuable given the difficulties involved in this problem. In this paper, we propose a novel IHT algorithm for LRTR, choosing sequential per-mode SVD truncation as its thresholding operator. This operator is less costly than those used in existing IHT algorithms for LRTR, and often leads to superior performance. Furthermore, by exploiting the sequential optimality of the employed modal projections, we derive recovery guarantees relying on restricted isometry constants. Though these guarantees are suboptimal, our numerical studies indicate that a quasi-optimal number of Gaussian measurements suffices for perfect data reconstruction. We also investigate a continuation technique which yields a sequence of progressively more complex estimated models until attaining a target mrank. When recovering real-world data, this strategy stabilizes the estimation error and can also accelerate convergence. In tensor completion, in particular, it can cope with nonideal characteristics of the sensed tensors and so is crucial for achieving a satisfactory performance. Extensive numerical experiments are reported, including the completion of hyperspectral imaging data and comparisons with several other existing approaches
The design of superhydrophobic stainless steel surfaces by controlling nanostructures: A key parameter to reduce the implantation of pathogenic bacteria
No full textReducing bacterial adhesion on substrates is fundamental for various industries. In this work, new superhydrophobic surfaces are created by electrodeposition of hydrophobic polymers (PEDOT-F-4 or PEDOT-H-8) on stainless steel with controlled topographical features, especially at a nano-scale. Results show that anti-bioadhesive and anti-biofilm properties require the control of the surface topographical features, and should be associated with a low adhesion of water onto the surface (Cassie-Baxter state) with limited crevice features at the scale of bacterial cells (nano-scale structures)
Shear reversal response of frictional suspensions
No full textInternational audienceLately, several studies pointed the key role of contact forces in the mechanical response of suspensions of non-Brownian rough spheres in Newtonian liquids. Both particle roughness [1] and frictional contacts [2] strongly increase the suspension viscosity. A Transition from frictionless to frictional contact between particles has been invoked to explain the discontinuous shear thickening in concentrated suspensions [3]. Recent numerical simulations [4] showed that shear reversal experiments, first proposed by Gadala-Maria and Acrivos [5], make it possible to estimate the contribution of contact forces to the viscosity. In this paper, we present the viscosity response to shear-reversal of suspensions made of either faceted sugar particles or PMMA spherical particles in a Newtonian silicon oil. In such a transient shear flow, the viscosity decreases after the shear inversion to a minimum where contacts vanish, and increases again up to the steady value as frictional contacts re-build [4,6].The steady viscosity and the minimum viscosity were measured for both suspensions as a function of particle volume fraction with the following results. (i) Steady viscosity are larger in sugar particle suspensions than in spherical particle suspensions. (ii) Both suspensions show close values of the minimum viscosity.We conclude from these experiments that contact forces between particles play a more important role in the steady rheological behavior of sugar particle suspensions than in spherical particle suspensions.[1] Tanner R. I. & Dai S. (2016). JOR 60, 809–818[2] Gallier S. et al. (2014). JFM 757, 514–549[3] Mari R et al. (2014). JOR 58, 1693–1724[4] Peters F et al. (2016). JOR 60, 715–732[5] Gadala-Maria F. & Acrivos A. (1980). JOR 24, 799-814[6] Lin N. Y. C. et al. (2015). PRL 115, 22830
Algorithmes de prix, intelligence artificielle et équilibres collusifs
No full textSciences Po OFCE Working Paper, n°14, 2017/05Les algorithmes de prix mis en œuvre par des firmes concurrentes peuvent constituer le support de collusions. Les ressources offertes par le Big Data, les possibilités d’ajustement des prix en temps réel et l’analyse prédictive peuvent permettre d’atteindre rapidement et de maintenir durablement des équilibres de collusion tacite. Le recours à l’intelligence artificielle pose un enjeu spécifique en ce sens que l’algorithme peut découvrir de lui-même l’intérêt d’un accord tacite de non-agression et que l’analyse de son processus décisionnel est particulièrement difficile. Ce faisant la sanction de l’entente sur la base du droit des pratiques anticoncurrentielles ne va pas de soi. L’article explore donc les voies de régulation possibles, que celles-ci passent par des audits ou par l’activation de règles de responsabilit
Overlapping community detection and temporal analysis on Q&A sites
No full textInternational audienceIn many social networks, people interact based on their relationship network. Community detection algorithms are then useful to reveal the sub-structures of a network. Identifying these users' communities can help us assist their life-cycle. However, in certain kinds of online communities such as question-and-answer (Q&A) sites or forums, people interact based on common topics of interest, rather than an explicit relationship network. Therefore, many traditional community detection techniques do not apply directly. Discovering those topics of interest is critical to identify users' communities. Besides, users' activities on certain topics of interest are evolving with time and it is therefore very important to extract their temporal dynamics. In this paper, we first propose Topic Trees Distributions (TTD), an efficient approach for extracting topics from Q&A sites in order to detect overlapping communities. We then extend TTD to propose Temporal Topic Expertise Activity (TTEA), a graphical probabilistic model to extract both topics-based expertise and temporal information. We evaluated and compared our models with state-of-the-art approaches on a dataset extracted from the popular Q&A site StackOverflow
Antenne à Résonateur Diélectrique pour réduire l’effet du corps humain dans un dispositif IoT miniature
No full textNational audienc
Compact Integrated IoT Antenna Robust to the Effects of the Environment
No full textInternational audienc
A Decoupling Technique for the Design of Strongly Isolated Closely Spaced Antennas
No full textInternational audienc
Convergence of the Chern-Moser-Beloshapka normal forms
No full textIn this article, we first describe a normal form of real-analytic, Levi-nondegenerate submanifolds of of codimension d ≥ 1 under the action of formal biholomorphisms, that is, of perturbations of Levi-nondegenerate hyperquadrics. We give a sufficient condition on the formal normal form that ensures that the normalizing transformation to this normal form is holomorphic. We show that our techniques can be adapted in the case d = 1 in order to obtain a new and direct proof of Chern-Moser normal form theorem
Evolutionary Analysis of the Mammalian Tuftelin Sequence Reveals Features of Functional Importance
No full textInternational audienceTuftelin (TUFT1) is an acidic, phosphorylated glycoprotein, initially discovered in developing enamel matrix. TUFT1 is expressed in many mineralized and non-mineralized tissues. We performed an evolutionary analysis of 82 mammalian TUFT1 sequences to identify residues and motifs that were conserved during 220 million years (Ma) of evolution. We showed that 168 residues (out of the 390 residues composing the human TUFT1 sequence) are under purifying selection. Our analyses identified several, new, putatively functional domains and confirmed previously described functional domains, such as the TIP39 interaction domain, which correlates with nuclear localization of the TUFT1 protein, that was demonstrated in several tissues. We also identified several sites under positive selection, which could indicate evolutionary changes possibly related to the functional diversification of TUFT1 during evolution in some lineages. We discovered that TUFT1 and MYZAP (myocardial zonula adherens protein) share a common ancestor that was duplicated circa 500 million years ago. Taken together, these findings expand our knowledge of TUFT1 evolution and provide new information that will be useful for further investigation of TUFT1 functions