Jurnal Edukasi
Not a member yet
4879 research outputs found
Sort by
Polynomial Template Generation using Sum-of-Squares Programming
No full text19 pages, 3 figures, 1 tableTemplate abstract domains allow to express more interesting properties than classical abstract domains. However, template generation is a challenging problem when one uses template abstract domains for program analysis. In this paper, we relate template computation with the program properties that we want to prove. We focus on one-loop programs with a conditional branch and assume that all the functions involved in these programs are polynomials. We formally define the notion of well-representative template basis with respect to a given property. The definition relies on the fact that template abstract domains produce inductive invariants. We show that these invariants can be obtained by solving certain systems of functional inequalities. Then, such systems can be strengthened using a hierarchy of sum-of-squares feasibility problems. Each step of the SOS hierarchy can possibly provide a solution which in turn yields feasible invariant bound together with a certificate that the desired property holds. The interest of this approach is illustrated on nontrivial program examples in polynomial arithmetic
New ferromagnetic core shapes for induction sensors
No full textInternational audienceInduction sensors are used in a wide range of scientific and industrial applications. One way to improve these is rigorous modelling of the sensor combined with a low voltage and current input noise preamplifier aiming to optimize the whole induction magnetometer. In this paper, we explore another way, which consists in the use of original ferromagnetic core shapes of induction sensors, which bring substantial improvements. These new configurations are the cubic, orthogonal and coiled-core induction sensors. For each of them we give modelling elements and discuss their benefits and drawbacks with respect to a given noise-equivalent magnetic induction goal. Our discussion is supported by experimental results for the cubic and orthogonal configurations, while the coiled-core configuration remains open to experimental validation. The transposition of these induction sensor configurations to other magnetic sensors (fluxgate and giant magneto-impedance) is an exciting prospect of this work
Excitation spectrum of the Lieb-Liniger model
No full textseven pages and two figuresInternational audienceWe study the integrable model of one-dimensional bosons with contact repulsion. In the limit of weak interaction, we use the microscopic hydrodynamic theory to obtain the excitation spectrum. The statistics of quasiparticles changes with the increase of momentum. At lowest momenta good quasiparticles are fermions, while at higher momenta they are Bogoliubov bosons, in accordance with recent studies. In the limit of strong interaction, we analyze the exact solution and find exact results for the spectrum in terms of the asymptotic series. Those results undoubtedly suggest that fermionic quasiparticle excitations actually exist at all momenta for moderate and strong interaction, and also at lowest momenta for arbitrary interaction. Moreover, at strong interaction we find highly accurate analytical results for several relevant quantities of the Lieb-Liniger model
Apprentissage de dictionnaire pour les représentations parcimonieuses
No full textThis is an abstract of the full preprint available at http://hal.inria.fr/hal-00918142/National audienceA popular approach within the signal processing and machine learning communities consists in modelling high-dimensional data as sparse linear combinations of atoms selected from a dictionary. Given the importance of the choice of the dictionary for the operational deployment of these tools, a growing interest for \emph{learned} dictionaries has emerged. The most popular dictionary learning techniques, which are expressed as large-scale matrix factorization through the optimization of a non convex cost function, have been widely disseminated thanks to extensive empirical evidence of their success and steady algorithmic progress. Yet, until recently they remained essentially heuristic. We will present recent work on statistical aspects of sparse dictionary learning, contributing to the characterization of the excess risk as a function of the number of training samples. The results cover non only sparse dictionary learning but also a much larger class of constrained matrix factorization problems.La modélisation de données de grande dimension comme combinaisons linéaires parcimonieuses d'atomes d'un dictionnaire est devenu un outil très populaire en traitement du signal et de l'image. Etant donné l'importance du choix du dictionnaire pour le déploiement opérationnel de ces outils, des approches basées sur l'apprentissage à partir d'une collection ont connu un bel essor. Les techniques les plus populaires abordent le problème sous l'angle de la factorisation de grandes matrices via la minimisation d'une fonction de coût non-convexe. Si des progrès importants en terme d'efficacité algorithmique ont favorisé leur diffusion, ces approches restaient jusqu'à récemment essentiellement empiriques. Nous présenterons des travaux récents abordant les aspects statistiques de ces techniques et contribuant à caractériser l'excès de risque en fonction du nombre d'exemples disponibles. Les résultats couvrent non seulement l'apprentissage de dictionnaire pour les représentations parcimonieuses, mais également une classe sensiblement plus large de factorisations de matrices sous contraintes
Phoretic self-propulsion at finite Peclet numbers
No full textInternational audiencePhoretic self-propulsion is a unique example of force- and torque-free motion on small scales. The classical framework describing the flow field around a particle swimming by self-diffusiophoresis neglects the advection of the solute field by the flow and assumes that the chemical interaction layer is thin compared to the particle size. In this paper we quantify and characterize the effect of solute advection on the phoretic swimming of a sphere. We first rigorously derive the regime of validity of the thin-interaction-layer assumption at finite values of the Péclet number (Pe). Under this assumption, we solve computationally the flow around Janus phoretic particles and examine the impact of solute advection on propulsion and the flow created by the particle. We demonstrate that although advection always leads to a decrease of the swimming speed and flow stresslet at high values of the Péclet number, an increase can be obtained at intermediate values of Pe. This possible enhancement of swimming depends critically on the nature of the chemical interactions between the solute and the surface. We then derive an asymptotic analysis of the problem at small Pe which allows us to rationalize our computational results. Our computational and theoretical analysis is accompanied by a parallel study of the influence of reactive effects at the surface of the particle (Damköhler number) on swimming
On using extreme values to detect global stability thresholds in multi-stable systems: The case of transitional plane Couette flow
No full textInternational audienceExtreme Value Theory (EVT) is exploited to determine the global stability threshold Rg of plane Couette flow - the flow of a viscous fluid in the space between two parallel plates - whose laminar or turbulent behavior depends on the Reynolds number R . Even if the existence of a global stability threshold has been detected in simulations and experiments, its numerical value has not been unequivocally defined. Rg is the value such that for R>Rg, turbulence is sustained, whereas for
Nitrile Synthesis through Catalyzed Cascades Involving Acid-Nitrile Exchange
No full textInternational audienceIrreversible acid-nitrile exchange reactions using both glutaronitrile and (phenylsulfonyl)acetonitrile may be catalyzed by Lewis acids. Whereas a cyclization towards imides displaces the equilibria in the reaction with dinitriles, a decarboxylation step is involved when using the (phenylsulfonyl)acetonitrile
Differential Interaction Kinetics of a Bipolar Structure- Specific Endonuclease with DNA Flaps Revealed by Single-Molecule Imaging Differential Interaction Kinetics of a Bipolar Structure-Specific Endonuclease with DNA Flaps
No full textInternational audienceAs DNA repair enzymes are essential for preserving genome integrity, understanding their substrate interaction dynamics and the regulation of their catalytic mechanisms is crucial. Using single-molecule imaging, we investigated the association and dissociation kinetics of the bipolar endonuclease NucS from Pyrococcus abyssi (Pab) on 59 and 39-flap structures under various experimental conditions. We show that association of the PabNucS with ssDNA flaps is largely controlled by diffusion in the NucS-DNA energy landscape and does not require a free 59 or 39 extremity. On the other hand, NucS dissociation is independent of the flap length and thus independent of sliding on the single-stranded portion of the flapped DNA substrates. Our kinetic measurements have revealed previously unnoticed asymmetry in dissociation kinetics from these substrates that is markedly modulated by the replication clamp PCNA. We propose that the replication clamp PCNA enhances the cleavage specificity of NucS proteins by accelerating NucS loading at the ssDNA/dsDNA junctions and by minimizing the nuclease interaction time with its DNA substrate. Our data are also consistent with marked reorganization of ssDNA and nuclease domains occurring during NucS catalysis, and indicate that NucS binds its substrate directly at the ssDNA-dsDNA junction and then threads the ssDNA extremity into the catalytic site. The powerful techniques used here for probing the dynamics of DNA-enzyme binding at the single-molecule have provided new insight regarding substrate specificity of NucS nucleases
An Eco-Evolutionary approach of Adaptation and Recombination in a large population of varying size
No full textAccepted in SPAWe identify the genetic signature of a selective sweep in a population described by a birth-and-death process with density dependent competition. We study the limit behaviour for large K, where K scales the population size. We focus on two loci: one under selection and one neutral. We distinguish a soft sweep occurring after an environmental change, from a hard sweep occurring after a mutation, and express the neutral proportion variation as a function of the ecological parameters, recombination probability r_K, and K. We show that for a hard sweep, two recombination regimes appear according to the order of r_K log K
LeakWatch: Estimating Information Leakage from Java Programs
No full textInternational audiencePrograms that process secret data may inadvertently reveal information about those secrets in their publicly-observable output. This paper presents LeakWatch, a quantitative information leakage analysis tool for the Java programming language; it is based on a flexible "point-to-point" information leakage model, where secret and publicly-observable data may occur at any time during a program's execution. LeakWatch repeatedly executes a Java program containing both secret and publicly-observable data and uses robust statistical techniques to provide estimates, with confidence intervals, for min-entropy leakage (using a new theoretical result presented in this paper) and mutual information. We demonstrate how LeakWatch can be used to estimate the size of information leaks in a range of real-world Java programs