Publications

2018

• The Crypto-Agility Properties
  
  • Aissaoui-Mehrez Hassane
  • El Omri Othmane
  , 2018.
• A Robust Dual Reference Computing-in-Memory Implementation and Design Space Exploration Within STT-MRAM
  
  • Zhang Liuyang
  • Kang Wang
  • Cai Hao
  • Ouyang Peng
  • Torres Lionel
  • Zhang Youguang
  • Todri-Sanial Aida
  • Zhao Weisheng
  , 2018, pp.275-280. Due to the "memory wall" in conventional Von-Neumann computer architectures, the limited bandwidth between processors and memories has become one of the most critical bottlenecks to improve system performance. With the emerging of non-volatile memories, the computing-in-memory (CIM) paradigm has regained interest to tackle the issue at the architecture level. CIM can effectively alleviate the stress on the limitted bandwidth by performing logic operations within memories. However, CIMs are not yet studied carefully at the circuit level, and even its reliability and performance. In this paper, we proposed a CIM implementation: dual reference (DualRef) scheme at the circuit level within STT-MRAM (Spin Transfer Torque Magnetic Random Access Memory) array. Simulations were carried out to verify the functionality and assess the reliability and performance of DualRef scheme in terms of operation error rate, sensing margin, operation delay and dynamic energy consumption. Simulation results validate DualRef scheme and reveal that it is reliable to perform bitwise logic opertions within STT-MRAM while the TMR (Tunnel Magnetoresistance Ratio) varying between 100% and 300% and supply voltage Vdd varying from 0.9V to 1.2V. This work provides a robust circuitry scheme and design space to effectively implement CIM in STT-MRAM. (10.1109/ISVLSI.2018.00058)
  DOI : 10.1109/ISVLSI.2018.00058
• Stein's method and Papangelou intensity for Poisson or Cox process approximation
  
  • Decreusefond Laurent
  • Vasseur Aurélien
  , 2018. In this paper, we apply the Stein's method in the context of point processes, namely when the target measure is the distribution of a finite Pois-son point process. We show that the so-called Kantorovich-Rubinstein distance between such a measure and another finite point process is bounded by the $L^1$-distance between their respective Papangelou intensities. Then, we deduce some convergence rates for sequences of point processes approaching a Poisson or a Cox point process.
• Jeu de données SemBib: représentation sémantique des données bibliographiques de Télécom ParisTech
  
  • Moissinac Jean-Claude Jc
  , 2018, pp.257-259. Nous allons présenter ici le jeu de données SemBib, représentation sémantique des données bibliographiques de Télécom ParisTech. Ce travail est mené dans le cadre du projet SemBib, au sein de Telecom ParisTech.
• A Survey of Simple Geometric Primitives Detection Methods for Captured 3D Data
  
  • Kaiser Adrien
  • Ybanez Zepeda Jose Alonso
  • Boubekeur Tamy
  Computer Graphics Forum, Wiley, 2018, 38 (1), pp.167-196. The amount of captured 3D data is continuously increasing, with the democratization of consumer depth cameras, the development of modern multi‐view stereo capture setups and the rise of single‐view 3D capture based on machine learning. The analysis and representation of this ever growing volume of 3D data, often corrupted with acquisition noise and reconstruction artefacts, is a serious challenge at the frontier between computer graphics and computer vision. To that end, segmentation and optimization are crucial analysis components of the shape abstraction process, which can themselves be greatly simplified when performed on lightened geometric formats. In this survey, we review the algorithms which extract simple geometric primitives from raw dense 3D data. After giving an introduction to these techniques, from the acquisition modality to the underlying theoretical concepts, we propose an application‐oriented characterization, designed to help select an appropriate method based on one's application needs and compare recent approaches. We conclude by giving hints for how to evaluate these methods and a set of research challenges to be explored. (10.1111/cgf.13451)
  DOI : 10.1111/cgf.13451
• Utilisation de fibres à couplage inhibé pour le controle de l'intrication spectrale de paires de photons
  
  • Cordier Martin
  • Orieux Adeline
  • Debord Benoît
  • Gérôme Frédéric
  • Gorse Alexandre
  • Chafer Matthieu
  • Diamanti Eleni
  • Delaye Philippe
  • Benabid Fetah
  • Zaquine Isabelle
  , 2018.
• Impact du canal sur les transmissions satellites-sol corrigées par optique adaptative en détection cohérente
  
  • Paillier Laurie
  • Conan Jean-Marc
  • Védrenne Nicolas
  • Le Bidan Raphaël
  • Artaud Géraldine
  • Jaouën Yves
  , 2018.
• Scalable models for points-of-interest recommender systems
  
  • Griesner Jean-Benoît
  , 2018. The task of points-of-interest (POI) recommendations has become an essential feature in location-based social networks. However it remains a challenging problem because of specific constraints of these networks. In this thesis I investigate new approaches to solve the personalized POI recommendation problem. Three main contributions are proposed in this work. The first contribution is a new matrix factorization model that integrates geographical and temporal influences. This model is based on a specific processing of geographical data. The second contribution is an innovative solution against the implicit feedback problem. This problem corresponds to the difficulty to distinguish among unvisited POI the actual "unknown" from the "negative" ones. Finally the third contribution of this thesis is a new method to generate recommendations with large-scale datasets. In this approach I propose to combine a new geographical clustering algorithm with users’ implicit social influences in order to define local and global mobility scales.
• Asynchronous Stochastic Quasi-Newton MCMC for Non-Convex Optimization
  
  • Şimşekli Umut
  • Yildiz Cagatay
  • Huy Nguyen Thanh
  • Richard Gael
  • Cemgil A Taylan
  , 2018. Recent studies have illustrated that stochastic gradient Markov Chain Monte Carlo techniques have a strong potential in non-convex optimization , where local and global convergence guarantees can be shown under certain conditions. By building up on this recent theory, in this study, we develop an asynchronous-parallel stochastic L-BFGS algorithm for non-convex optimization. The proposed algorithm is suitable for both distributed and shared-memory settings. We provide formal theoretical analysis and show that the proposed method achieves an ergodic convergence rate of O(1/ √ N) (N being the total number of iterations) and it can achieve a linear speedup under certain conditions. We perform several experiments on both synthetic and real datasets. The results support our theory and show that the proposed algorithm provides a significant speedup over the recently proposed synchronous distributed L-BFGS algorithm.
• Non réciprocité de l'émission des paires de photons dans les milieux non linéaires non uniformes
  
  • Harlé Thibault
  • Cordier Martin
  • Zaquine Isabelle
  • Delaye Philippe
  , 2018.
• Statistical analysis and modeling of soft or flexible antennas in fluctuating conditions
  
  • Du Jinxin
  , 2018. With the massive deployment of flexible electronics in the context of Internet of Things (IoT) and wireless body area networks (WBANs), the demand for soft or flexible antennas is substantially growing. Unlike traditional nondeformable or clearly isolated antennas whose characteristics are deterministic, soft/flexible antennas, especially when they are implemented in fluctuating conditions, are subject to various types of variability by essence random: geometric deformations, variations in the material properties, interactions with the surrounding environment (e.g. the human body), etc. These random uncertainties can significantly disturb the in situ performance of antennas and must be taken into account quantitatively. The characterization of these random effects cannot be effectively achieved using the conventional "electromagnetic simulation - prototyping - measurement" approach because of its very high cost (for each of these steps). In this context, surrogate models which can effectively predict these random effects with good precision are extremely useful. This dissertation focuses on the investigation of efficient methodologies dedicated to the construction of surrogate models for antennas undergoing random uncertainties. Two main objectives are targeted. On the one hand, describe quantitatively and parsimoniously the overall behavior of the antennas, that is to say, the complete radiated far-field (FF) as well as the reflection coefficient. On the other hand, propose efficient, reliable and versatile methodologies for the extraction of antenna models from a limited number of pre-evaluations. In particular, a statistical methodology based on polynomial chaos has been proposed and its effectiveness has been demonstrated by applying it firstly to a canonical antenna undergoing various types of deformations, then in a more realistic application framework, to an antenna printed on a fabric substrate. The benefit of the surrogate models thus developed is to have access to the responses of the "random antennas" at a very low calculation cost. From the application point of view, such surrogate models can be used in various types of higher-level analysis such as joint antenna/channel modeling, radio links analysis, antenna plugged-in asymptotic simulators (e.g. ray launching or tracing), Multi-Input & Multi-Output (MIMO) systems characterization, or beamforming.
• Multichannel Audio Modeling with Elliptically Stable Tensor Decomposition
  
  • Fontaine Mathieu
  • Stöter Fabian-Robert
  • Liutkus Antoine
  • Şimşekli Umut
  • Serizel Romain
  • Badeau Roland
  , 2018, 10891, pp.13-23. This paper introduces a new method for multichannel speech enhancement based on a versatile modeling of the residual noise spec-trogram. Such a model has already been presented before in the single channel case where the noise component is assumed to follow an alpha-stable distribution for each time-frequency bin, whereas the speech spec-trogram, supposed to be more regular, is modeled as Gaussian. In this paper, we describe a multichannel extension of this model, as well as a Monte Carlo Expectation-Maximisation algorithm for parameter estimation. In particular, a multichannel extension of the Itakura-Saito nonnegative matrix factorization is exploited to estimate the spectral parameters for speech, and a Metropolis-Hastings algorithm is proposed to estimate the noise contribution. We evaluate the proposed method in a challenging multichannel denoising application and compare it to other state-of-the-art algorithms. (10.1007/978-3-319-93764-9_2)
  DOI : 10.1007/978-3-319-93764-9_2
• Securing Delay-Tolerant IoT Uplink Communications Against Eavesdropping
  
  • Iellamo Stefano
  • Guiazon Raoul
  • Coupechoux Marceau
  • Wong Kai-Kit
  , 2018, pp.1-8. We consider a network of Internet of Things devices transmitting to an IoT Gateway (IoT-GW). Such communications can potentially be overheard by one or multiple eavesdroppers. Our goal is to design an artificial noise (AN)-aided transmit strategy in order to enhance security against eavesdropping. We propose a communication design where the potential eavesdrop- pers are deactivated by means of jamming operations performed by 1) an In-Band Full Duplex (IBFD) IoT-GW and/or by 2) cooperative helpers featuring multiple antennas. We show that the solution where only the IBFD IoT-GW generates AN is feasible for small IoT networks and when a neutralization zone around each IoT-device is assumed. In the case with helpers instead, we show that the Average number of Secure Connections (ASC) increases at least exponentially with the density of the helpers. (10.1109/CIOT.2018.8627099)
  DOI : 10.1109/CIOT.2018.8627099
• Formal and Virtual Multi-level Design Space Exploration.
  
  • Li Letitia W.
  • Genius Daniela
  • Apvrille Ludovic
  , 2018, pp.47-71. With the growing complexity of embedded systems, a systematic design process and tool are vital to help designers assure that their design meets specifications. The design of an embedded system evolves through multiple modeling phases, with varying levels of abstraction. A modeling toolkit should also support the various evaluations needed at each stage, in the form of simulation, formal verification, and performance evaluation. This chapter introduces our model-based engineering process with the supporting toolkit TTool, with two main design stages occurring at a different level of abstraction. A system-level design space exploration selects the architecture and partitions functions into hardware and software. The subsequent software design phase then designs and assesses the detailed functionality of the system, and evaluates the partitioning choices. We illustrate the design phases and supported evaluations with a Smart Card case study. (10.1007/978-3-319-94764-8_3)
  DOI : 10.1007/978-3-319-94764-8_3
• Conception d'un filtre optique à profil arbitraire de signaux micro-ondes par dffusion Brillouin stimulée
  
  • Wei Wei
  • Jaouën Yves
  • Yi L. L.
  • Hu W.
  , 2018, pp.Session O2-B.
• Implementation Security of Quantum Cryptography: Introduction, challenges, solutions
  
  • Alleaume Romain
  ETSI White Paper, 2018, 27, pp.28.
• A line segment detector for SAR images with controlled false alarm rate
  
  • Liu Chenguang
  • Abergel Rémy
  • Gousseau Yann
  • Tupin Florence
  , 2018. In this paper we propose to adapt LSD [1] (a state-of-the-art line segment detector for optical images) to SAR images. The first modification is replacing the gradient computation with an exponentially weighted ratio-based method which has a constant false alarm rate for SAR images. Next, we observe that the strong noise removal necessary for processing SAR images strongly impairs the independent hypothesis of the a contrario model used by LSD. A first order Markov chain is used to take the spatial dependencies into consideration. Experiments show that the proposed method has good perfor- mances and the number of false detections is well controlled.
• Bounds on the approximation power of feedforward neural networks
  
  • Mehrabi Mohammad
  • Tchamkerten Aslan
  • Yousefi Mansoor
  , 2018.
• Sample-and-hold circuit for an electrical signal
  
  • Meyer Arnaud
  • Louis Bruno
  • Corbière Rémi
  • Petit Vincent
  • Desgreys Patricia
  • Petit Hervé
  , 2018.
• Exploiting Contextual and External Data for Hotel Recommendation
  
  • Al-Ghossein Marie
  • Abdessalem Talel
  • Barré Anthony
  , 2018, pp.323-328. (10.1145/3213586.3225245)
  DOI : 10.1145/3213586.3225245
• A Conditional Gradient Framework for Composite Convex Minimization with Applications to Semidefinite Programming
  
  • Yurtsever Alp
  • Fercoq Olivier
  • Cevher Volkan
  • Locatello Francesco
  , 2018. We propose a conditional gradient framework for a composite convex minimization template with broad applications. Our approach combines the notions of smoothing and homotopy under the CGM framework, and provably achieves the optimal O(1/√k) convergence rate. We demonstrate that the same rate holds if the linear subproblems are solved approximately with additive or multiplicative error. Specific applications of the framework include the non-smooth minimization, semidefinite programming, and minimization with linear inclusion constraints over a compact domain. We provide numerical evidence to demonstrate the benefits of the new framework.
• Représentation des territoires dans des grandes bases de connaissances comme DBpedia et Yago
  
  • Moissinac Jean-Claude Jc
  , 2018.
• Network-level strategies for best use of optical functionalities
  
  • Minakhmetov Artur
  • Chouman Hussein
  • Iannone Luigi
  • Lourdiane Mounia
  • Ware Cédric
  , 2018 (Tu.B1.3).
• Learning Kolmogorov Models for Binary Random Variables
  
  • Ghauch Hadi
  • Skoglund Mikael
  • Shokri-Ghadikolaei Hossein
  • Fischione Carlo
  • Sayed Ali
  , 2018. We summarize our recent findings Authors (2017), where we proposed a framework for learning a Kolmogorov model, for a collection of binary random variables. More specifically, we derive conditions that causally link outcomes of specific random variables, and extract valuable relations from the data. We also propose an algorithm for computing the model and show its first-order optimality, despite the combinatorial nature of the learning problem. We apply the proposed algorithm to recommendation systems, although it is applicable to other scenarios. We believe that the work is a significant step toward interpretable machine learning.
• Deterministic Core Scrambling for Multi-Core Fiber Transmission
  
  • Abouseif Akram
  • Rekaya-Ben Othman Ghaya
  • Jaouën Yves
  , 2018, paper 5B2-4.