Search

Unsupervised domain mapping aims to learn a function to translate domain X to Y by a function GXY in the absence of paired examples. Finding the optimal GXY without paired data is an ill-posed problem, so appropriate constraints are required to obtain reasonable solutions. One of the most prominent constraints is cycle consistency, which enforces the translated image by GXY to be translated back to the input image by an inverse mapping GYX. While cycle consistency requires the simultaneous training of GXY and GY X, recent studies have shown that one-sided domain mapping can be achieved by preserving pairwise distances between images. Although cycle consistency and distance preservation successfully constrain the solution space, they overlook the special properties that simple geometric transformations do not change the semantic structure of images. Based on this special property, we develop a geometry-consistent generative adversarial network (GcGAN), which enables one-sided unsupervised domain mapping. GcGAN takes the original image and its counterpart image transformed by a predefined geometric transformation as inputs and generates two images in the new domain coupled with the corresponding geometry-consistency constraint. The geometry-consistency constraint reduces the space of possible solutions while keep the correct solutions in the search space. Quantitative and qualitative comparisons with the baseline (GAN alone) and the state-of-the-art methods including CycleGAN and DistanceGAN demonstrate the effectiveness of our method.

International audience; We propose a new approach for outdoor large scale image based localization that can deal with challenging scenarios like cross-season, cross-weather, day/night and long-term localization. The key component of our method is a new learned global image descriptor, that can effectively benefit from scene geometry information during training. At test time, our system is capable of inferring the depth map related to the query image and use it to increase localization accuracy. We are able to increase recall@1 performances by 2.15% on cross-weather and long-term localization scenario and by 4.24% points on a challenging winter/summer localization sequence versus state-of-the-art methods. Our method can also use weakly annotated data to localize night images across a reference dataset of daytime images.

We present precise photospheric parameters of 282 M dwarfs determined from fitting the most recent version of PHOENIX models to high-resolution CARMENES spectra in the visible (0.52-0.96 μm) and NIR wavelength range (0.96-1.71 μm). With its aim to search for habitable planets around M dwarfs, several planets of different masses have been detected. The characterization of the target sample is important for the ability to derive and constrain the physical properties of any planetary systems that are detected. As a continuation of previous work in this context, we derived the fundamental stellar parameters effective temperature, surface gravity, and metallicity of the CARMENES M-dwarf targets from PHOENIX model fits using a χ method. We calculated updated PHOENIX stellar atmosphere models that include a new equation of state to especially account for spectral features of lowerature stellar atmospheres as well as new atomic and molecular line lists. We show the importance of selecting magnetically insensitive lines for fitting to avoid effects of stellar activity in the line profiles. For the first time, we directly compare stellar parameters derived from multiwavelength range spectra, simultaneously observed for the same star. In comparison with literature values we show that fundamental parameters derived from visible spectra and visible and NIR spectra combined are in better agreement than those derived from the same spectra in the NIR alone.© 2019 ESO., CARMENES is funded by the German Max-Planck-Gesellschaft (MPG), the Spanish Consejo Superior de Investigaciones Cientificas (CSIC), the European Union through FEDER/ERF FICTS-2011-02 funds, and the members of the CARMENES Consortium (Max-Planck-Institut fur Astronomie, Instituto de Astrofisica de Andalucia, Landessternwarte Konigstuhl, Institut de Ciencies de l'Espai, Insitut fur Astrophysik Gottingen, Universidad Complutense de Madrid, Thuringer Landessternwarte Tautenburg, Instituto de Astrofisica de Canarias, Hamburger Sternwarte, Centro de Astrobiologia, and Centro Astronomico Hispano-Aleman), with additional contributions by the Spanish Ministry of Economy, the German Science Foundation through the Major Research Instrumentation Programme, and DFG Research Unit FOR2544 >Blue Planets around Red Stars>, the Klaus Tschira Stiftung, the states of Baden-Wurttemberg and Niedersachsen, and by the Junta de Andalucia. This work is based on data from the CARMENES data archive at CAB (INTA-CSIC). We acknowledge financial support from the Agencia Estatal de Investigacion of the Ministerio de Ciencia, Innovacion y Universidades, and the European FEDER/ERF funds through projects AYA2018-84089, ESP2017-87676-C5-1-R, ESP2016-80435-C2-1-R, AYA2016-79425-C3-1/2/3-P, AYA2015-69350-C3-2-P, and the Centre of Excellence > Severo Ochoa> and >Maria de Maeztu> awards to the Instituto de Astrofisica de Canarias (SEV-2015-0548), Instituto de Astrofisica de Andalucia (SEV-2017-0709), and Centro de Astrobiologia (MDM-2017-0737), and the Generalitat de Catalunya/CERCA programme. Some of the calculations presented here were performed at the RRZ of the Universitat Hamburg, at the Hochstleistungs Rechenzentrum Nord (HLRN), and at the National Energy Research Supercomputer Center (NERSC), which is supported by the Office of Science of the U.S. Department of Energy under Contract No. DE-AC03-76SF00098. We thank all these institutions for a generous allocation of computer time. P. H. H. gratefully acknowledges the support of NVIDIA Corporation with the donation of a Quadro P6000 GPU used in this research. This work has made use of the VALD database, operated at Uppsala University, the Institute of Astronomy RAS in Moscow, and the University of Vienna. This work has made use of data from the European Space Agency (ESA) mission Gaia (https://www.cosmos.esa.int/gaia), processed by the Gaia Data Processing and Analysis Consortium (DPAC, https://www.cosmos.esa.int/web/gaia/dpac/consortium).Funding for the DPAC has been provided by national institutions, in particular the institutions participating in the Gaia Multilateral Agreement.

International audience; Compressed sensing allows the acquisition of compressible signals with a small number of measurements. In experimental settings, the sensing process corresponding to the hardware implementation is not always perfectly known and may require a calibration. To this end, blind calibration proposes to perform at the same time the calibration and the compressed sensing. Schülke and collaborators suggested an approach based on approximate message passing for blind calibration (cal-AMP) in [1, 2]. Here, their algorithm is extended from the already proposed offline case to the online case, for which the calibration is refined step by step as new measured samples are received. We show that the performance of both the offline and the online algorithms can be theoretically studied via the State Evolution (SE) formalism. Finally, the efficiency of cal-AMP and the consistency of the theoretical predictions are confirmed through numerical simulations.

International audience; The ATIS (Air Travel Information Service) corpus will be soon celebrating its 30th birthday. Designed originally to benchmark spoken language systems, it still represents the most well-known corpus for benchmarking Spoken Language Understanding (SLU) systems. In 2010, in a paper titled "What is left to be understood in ATIS?" [1], Tur et al. discussed the relevance of this corpus after more than 10 years of research on statistical models for performing SLU tasks. Nowadays, in the Deep Neural Network (DNN) era, ATIS is still used as the main benchmark corpus for evaluating all kinds of DNN models, leading to further improvements, although rather limited, in SLU accuracy compared to previous state-of-the-art models. We propose in this paper to investigate these results obtained on ATIS from a qualitative point of view rather than just a quantitative point of view and answer the two following questions: what kind of qualitative improvement brought DNN models to SLU on the ATIS corpus? Is there anything left, from a qualitative point of view, in the remaining 5% of errors made by current state-of-the-art models?

International audience; Near-eye (VR/AR) displays suffer from technical, interaction as well as visual quality issues which hinder their commercial potential. This tutorial will deliver an overview of cutting-edge VR/AR display technologies, focusing on technical, interaction and perceptual issues which, if solved, will drive the next generation of display technologies. The most recent advancements in near-eye displays will be presented providing (i) correct accommodation cues, (ii) near-eye varifocal AR, (iii) high dynamic range rendition, (iv) gaze-aware capabilities, either predictive or based on eye-tracking as well as (v) motion-awareness. Future avenues for academic and industrial research related to the next generation of AR/VR display technologies will be analyzed.

Convolutional neural networks have been widely used in medical image segmentation. The amount of training data strongly determines the overall performance. Most approaches are applied for a single imaging modality, e.g., brain MRI. In practice, it is often difficult to acquire sufficient training data of a certain imaging modality. The same anatomical structures, however, may be visible in different modalities such as major organs on abdominal CT and MRI. In this work, we investigate the effectiveness of learning from multiple modalities to improve the segmentation accuracy on each individual modality. We study the feasibility of using a dual-stream encoder-decoder architecture to learn modality-independent, and thus, generalisable and robust features. All of our MRI and CT data are unpaired, which means they are obtained from different subjects and not registered to each other. Experiments show that multi-modal learning can improve overall accuracy over modality-specific training. Results demonstrate that information across modalities can in particular improve performance on varying structures such as the spleen.

Thesis submitted to the University of the Basque Country for the degree of Doctor in Physics., La capacidad de entender la materia y su interacción con el medio determinan en gran parte el desarrollo de las nuevas tecnologías. Hoy en día, las nanotecnologías son un campo emergente de investigación debido al gran impacto que tienen en la sociedad. Las simulaciones computacionales de fenómenos físicos en la nanoescala han contribuido a la aceleración de su desarrollo. En esta tesis doctoral nos hemos basado en simulaciones ab initio atomísticas para explicar el comportamiento de nanopartículas sometidas a estímulos externos. Con “atomístico” nos referimos a que la geometría del sistema se descibe mediante posiciones realístas de los átomos, es decir, que se tienen en cuenta las posiciones atómicas y la atracción Coulombiana generada por cada núcleo, en vez de reemplazarlas por un potencial efectivo suave que confina los electrónes en objetos de forma simple, como una esfera. “ab initio” significa que nos hemos basado en las leyes de la mecánica cuántica para modelar los electrones del sistema. De esta forma, en este trabajo hemos podido simular la interacción de centenas de electrones confinados dentro de nanopartículas, tanto entre ellos como con el medio. Este problema se conoce como el problema de muchos cuerpos ( many body problem en inglés). Desafortunadamente, hay que recurrir a aproximaciones para resolver el problema de muchos cuerpos. Las aproximaciones adoptadas en esta tesis son las integradas en la teoriá del funcional de la densidad (DFT, en inglés density functional theory) implementadas en los paquetes de SIESTA (Spanish Initiative for Electronic Simulations with Thousands of Atoms), así como su extensión a fenómenos dependientes del tiempo (TDDFT, en inglés timedependent DFT) implementadas en los paquetes de MBPT-LCAO (Many Body Perturbation Theory - Linear Combination of Atomic Orbitals) y PySCFNAO (Python-Based Simulations of Chemistry Framework - Numerical Atomic Orbitals). Por otra parte, en esta tesis doctoral hemos conseguido implementar con éxito nuevas funcionalidades en los paquetes de MBPT-LCAO y PySCF-NAO. Ambos paquetes han sido utilizados para calcular las propiedades de estados excitados de sistemas finitos como agregados atómicos (clusters en inglés) metálicos, fragmentos de nanotubos y moléculas pequeñas., I have to acknowledge the financial support from the Departamento de Educación of the Basque Government through a PhD grant, as well as from Euskampus and the DIPC at the initial stage of my work. I gratefully acknowledge the support of NVIDIA Corporation with the donation of the Tesla K40 GPU used for this research.

Explainability and interpretability are two critical aspects of decision support systems. Within computer vision, they are critical in certain tasks related to human behavior analysis such as in health care applications. Despite their importance, it is only recently that researchers are starting to explore these aspects. This paper provides an introduction to explainability and interpretability in the context of computer vision with an emphasis on looking at people tasks. Specifically, we review and study those mechanisms in the context of first impressions analysis. To the best of our knowledge, this is the first effort in this direction. Additionally, we describe a challenge we organized on explainability in first impressions analysis from video. We analyze in detail the newly introduced data set, the evaluation protocol, and summarize the results of the challenge. Finally, derived from our study, we outline research opportunities that we foresee will be decisive in the near future for the development of the explainable computer vision field., Comment: Preprint submitted to TAC

We present DLTK, a toolkit providing baseline implementations for efficient experimentation with deep learning methods on biomedical images. It builds on top of TensorFlow and its high modularity and easy-to-use examples allow for a low-threshold access to state-of-the-art implementations for typical medical imaging problems. A comparison of DLTK's reference implementations of popular network architectures for image segmentation demonstrates new top performance on the publicly available challenge data "Multi-Atlas Labeling Beyond the Cranial Vault". The average test Dice similarity coefficient of $81.5$ exceeds the previously best performing CNN ($75.7$) and the accuracy of the challenge winning method ($79.0$).

Deep learning approaches such as convolutional neural nets have consistently outperformed previous methods on challenging tasks such as dense, semantic segmentation. However, the various proposed networks perform differently, with behaviour largely influenced by architectural choices and training settings. This paper explores Ensembles of Multiple Models and Architectures (EMMA) for robust performance through aggregation of predictions from a wide range of methods. The approach reduces the influence of the meta-parameters of individual models and the risk of overfitting the configuration to a particular database. EMMA can be seen as an unbiased, generic deep learning model which is shown to yield excellent performance, winning the first position in the BRATS 2017 competition among 50+ participating teams., Comment: The method won the 1st-place in the Brain Tumour Segmentation (BRATS) 2017 competition (segmentation task)

International audience; The papers in this special issue focus on computer arithmetic which is used in many applications, usually totally silently (one should keep in mind that even when running programs that are not at all numeric, memory addresses are computed, which involves additions, multiplications, and sometimes divisions). However, in some areas, it plays a central role.

International audience; In this paper we propose to generalize Jaccard and related measures, often used as similarity coefficients between two sets. We define Jaccard, Dice-Sorensen and Tversky edge weights on a graph and generalize them to account for vertex weights. We develop an efficient parallel algorithm for computing Jaccard edge and PageRank vertex weights. We highlight that the weights computation can obtain more than 10× speedup on the GPU versus CPU on large realistic data sets. Also, we show that finding a minimum balanced cut for modified weights can be related to minimizing the sum of ratios of the intersection and union of nodes on the boundary of clusters. Finally, we show that the novel weights can improve the quality of the graph clustering by about 15% and 80% for multi-level and spectral graph partitioning and clustering schemes, respectively.

In this paper we develop a parallel approach for computing the modularity clustering often used to identify and analyse communities in social networks. We show that modularity can be approximated by looking at the largest eigenpairs of the weighted graph adjacency matrix that has been perturbed by a rank one update. Also, we generalize this formulation to identify multiple clusters at once. We develop a fast parallel implementation for it that takes advantage of the Lanczos eigenvalue solver and k-means algorithm on the GPU. Finally, we highlight the performance and quality of our approach versus existing state-of-the-art techniques.

Significant advances have been made towards building accu- rate automatic segmentation systems for a variety of biomedical applica- tions using machine learning. However, the performance of these systems often degrades when they are applied on new data that differ from the training data, for example, due to variations in imaging protocols. Man- ually annotating new data for each test domain is not a feasible solution. In this work we investigate unsupervised domain adaptation using ad- versarial neural networks to train a segmentation method which is more invariant to differences in the input data, and which does not require any annotations on the test domain. Specifically, we learn domain-invariant features by learning to counter an adversarial network, which attempts to classify the domain of the input data by observing the activations of the segmentation network. Furthermore, we propose a multi-connected domain discriminator for improved adversarial training. Our system is evaluated using two MR databases of subjects with traumatic brain in- juries, acquired using different scanners and imaging protocols. Using our unsupervised approach, we obtain segmentation accuracies which are close to the upper bound of supervised domain adaptation.

Well-suited strategy for GPU computing of ESO method driven by isosurfaces is proposed.Comparison of GPU instance of PCG using Jacobi and geometric multigrid preconditioning is presented.Different granularities are used to balance the workload of CUDA threads.Significant speedups with respect to sparse-matrix CPU implementation are achieved.High resolution truss-like designs of real-world topology optimization problems are presented. Evolutionary topology optimization of three-dimensional continuum structures is a computationally demanding task in terms of memory consumption and processing time. This work aims to alleviate these constraints proposing a well-suited strategy for Graphics Processing Unit (GPU) computing. Such a proposal adopts a fine-grained GPU instance of matrix-free iterative solver for structural analysis and an efficient GPU implementation for isosurface extraction and volume fraction calculation. The performance of the solving stage is evaluated using two preconditioning techniques, including the comparison with the sparse-matrix CPU implementation. The proposal is evaluated using topology optimization problems for real-world applications.

International audience; Automatic continuous affect recognition from audiovisual cues is arguably one of the most active research areas in machine learning. In addressing this regression problem, the advantages of the models, such as the global-optimisation capability of Support Vector Machine for Regression and the context-sensitive capability of memory-enhanced neural networks, have been frequently explored, but in an isolated way. Motivated to leverage the individual advantages of these techniques, this paper proposes and explores a novel framework, Strength Modelling, where two models are concatenated in a hierarchical framework. In doing this, the strength information of the first model, as represented by its predictions, is joined with the original features, and this expanded feature space is then utilised as the input by the successive model. A major advantage of Strength Modelling, besides its ability to hierarchically explore the strength of different machine learning algorithms, is that it can work together with the conventional feature-and decision-level fusion strategies for multimodal affect recognition. To highlight the effectiveness and robustness of the proposed approach, extensive experiments have been carried out on two time-and value-continuous spontaneous emotion databases (RECOLA and SEMAINE) using audio and video signals. The experimental results indicate that employing Strength Modelling can deliver a significant performance improvement for both arousal and valence in the unimodal and bimodal settings. The results further show that the proposed systems is competitive or outperform the other state-of-the-art approaches, but being with a simple implementation.

International audience; During the last decade, speech emotion recognition technology has matured well enough to be used in some real-life scenarios. However, these scenarios require an almost silent environment to not compromise the performance of the system. Emotion recognition technology from speech thus needs to evolve and face more challenging conditions, such as environmental additive and convolutional noises, in order to broaden its applicability to real-life conditions. This contribution evaluates the impact of a front-end feature enhancement method based on an autoencoder with long short-term memory neural networks, for robust emotion recognition from speech. Support Vector Regression is then used as a back-end for time- and value-continuous emotion prediction from enhanced features. We perform extensive evaluations on both non-stationary additive noise and convolutional noise, on a database of spontaneous and natural emotions. Results show that the proposed method significantly outperforms a system trained on raw features, for both arousal and valence dimensions, while having almost no degradation when applied to clean speech.

Part 6: Software Vulnerabilities; International audience; Despite the emphasis on building secure software, the number of vulnerabilities found in our systems is increasing every year, and well-understood vulnerabilities continue to be exploited. A common response to vulnerabilities is patch-based mitigation, which does not completely address the flaw and is often circumvented by an adversary. The problem actually lies in a lack of understanding of the nature of vulnerabilities. Vulnerability taxonomies have been proposed, but their usability is limited because of their ambiguity and complexity. This paper presents a taxonomy that views vulnerabilities as fractures in the interpretation of information as it flows in the system. It also presents a machine learning study validating the taxonomy’s unambiguity. A manually labeled set of 641 vulnerabilities trained a classifier that automatically categorized more than 70000 vulnerabilities from three distinct databases with an average success rate of 80 %. Important lessons learned are discussed such as (i) approximately 12 % of the studied reports provide insufficient information about vulnerabilities, and (ii) the roles of the reporter and developer are not leveraged, especially regarding information about tools used to find vulnerabilities and approaches to address them.

International audience

International audience; In this article, we propose a parallel implementation of the dynamic programming method for the knapsack problem on NVIDIA GPU. A GTX 260 (192 cores, 1.4GHz) was used for computational tests and processing times obtained with the parallel code are compared to the sequential one on a CPU with an Intel Xeon 3.0GHz. The results show a speedup up to 26 and permit one to solve large size problems within a reasonable processing time. Furthermore, in order to limit the communication between the CPU and the GPU, a compression technique is presented which decreases significantly the memory occupancy.

This Open Access Article is licensed under a Creative Commons Attribution 3.0 Unported Licence., Real-space grids are a powerful alternative for the simulation of electronic systems. One of the main advantages of the approach is the flexibility and simplicity of working directly in real space where the different fields are discretized on a grid, combined with competitive numerical performance and great potential for parallelization. These properties constitute a great advantage at the time of implementing and testing new physical models. Based on our experience with the Octopus code, in this article we discuss how the real-space approach has allowed for the recent development of new ideas for the simulation of electronic systems. Among these applications are approaches to calculate response properties, modeling of photoemission, optimal control of quantum systems, simulation of plasmonic systems, and the exact solution of the Schrödinger equation for low-dimensionality systems., XA acknowledges that part of this work was performed under the auspices of the U.S. Department of Energy at Lawrence Livermore National Laboratory under Contract DE-AC52-07A27344. XA and AA-G would like to thank the support received from Nvidia Corporation through the CUDA Center of Excellence program and the US Defense Threat Reduction Agency under contract no. HDTRA1-10-1-0046. DAS acknowledges support from the U.S. National Science Foundation graduate research program and IGERT fellowships, and from ARPA-E under grant DE-AR0000180. MJTO acknowledges financial support from the Belgian FNRS through FRFC project number 2.4545.12 “Control of attosecond dynamics: applications to molecular reactivity”. NH and IT received support from a Emmy-Noether grant from Deutsche Forschungsgemeinschaft. JAR, AV, UDG, AHL and AR ackowledge financial support by the European Research Council Advanced Grant DYNamo (ERC-2010-AdG-267374), European Commission project CRONOS (Grant number 280879-2 CRONOS CP-FP7), Marie Curie ITN POCAONTAS (FP7-PEOPLE-2012-ITN, project number 316633), OST Actions CM1204 (XLIC), and MP1306 (EUSpec), Spanish Grant (FIS2013-46159-C3-1-P) and Grupo Consolidado UPV/EHU del Gobierno Vasco (IT578-13). JAR acknowledges the Department of Education, Universities and Research of the Basque Government (grant IT395-10). AC acknowledges support from the grant FIS2013-46159-C3-2-P of the Spanish government.

International audience; A key concern in the design of controllers in wireless sensor network (WSN) nodes is the flexibility to execute different control tasks involving sensing, communications and computational resources of the node. In this paper, low power flexible controllers for WSN nodes based on reconfigurable microtasks composed of an FSM and datapath are presented. Coarse grain power gating opportunities are exploited in FSM and datapath for low power operation in reconfigurable microtasks. Power estimation results on typical benchmark microtasks show a 2× to 5× improvement in energy efficiency w.r.t a microcontroller at a cost of 5× relative to a microtask implemented as an ASIC with higher NRE costs.

The use of atomistic methods, such as the Continuous Cellular Automaton (CCA), is currently regarded as a computationally efficient and experimentally accurate approach for the simulation of anisotropic etching of various substrates in the manufacture of Micro-electro-mechanical Systems (MEMS). However, when the features of the chemical process are modified, a time-consuming calibration process needs to be used to transform the new macroscopic etch rates into a corresponding set of atomistic rates. Furthermore, changing the substrate requires a labor-intensive effort to reclassify most atomistic neighborhoods. In this context, the Level Set (LS) method provides an alternative approach where the macroscopic forces affecting the front evolution are directly applied at the discrete level, thus avoiding the need for reclassification and/or calibration. Correspondingly, we present a fully-operational Sparse Field Method (SFM) implementation of the LS approach, discussing in detail the algorithm and providing a thorough characterization of the computational cost and simulation accuracy, including a comparison to the performance by the most recent CCA model. We conclude that the SFM implementation achieves similar accuracy as the CCA method with less fluctuations in the etch front and requiring roughly 4 times less memory. Although SFM can be up to 2 times slower than CCA for the simulation of anisotropic etchants, it can also be up to 10 times faster than CCA for isotropic etchants. In addition, we present a parallel, GPU-based implementation (gSFM) and compare it to an optimized, multicore CPU version (cSFM), demonstrating that the SFM algorithm can be successfully parallelized and the simulation times consequently reduced, while keeping the accuracy of the simulations. Although modern multicore CPUs provide an acceptable option, the massively parallel architecture of modern GPUs is more suitable, as reflected by computational times for gSFM up to 7.4 times faster than for cSFM. © 2013 Elsevier B.V. All rights reserved., This work has been supported by the Spanish FPI-MICINN BES-2011-045940 grant and the Ramón y Cajal Fellowship Program by the Spanish Ministry of Science and Innovation. Also, we acknowledge support by the JAE-Doc grant from the Junta para la Ampliación de Estudios program co-funded by FSE and the Professor Partnership Program by NVIDIA Corporation.

Octopus is a general-purpose density-functional theory (DFT) code, with a particular emphasis on the time-dependent version of DFT (TDDFT). In this paper we present the ongoing efforts to achieve the parallelization of octopus. We focus on the real-time variant of TDDFT, where the time-dependent Kohn–Sham equations are directly propagated in time. This approach has great potential for execution in massively parallel systems such as modern supercomputers with thousands of processors and graphics processing units (GPUs). For harvesting the potential of conventional supercomputers, the main strategy is a multi-level parallelization scheme that combines the inherent scalability of real-time TDDFT with a real-space grid domain-partitioning approach. A scalable Poisson solver is critical for the efficiency of this scheme. For GPUs, we show how using blocks of Kohn–Sham states provides the required level of data parallelism and that this strategy is also applicable for code optimization on standard processors. Our results show that real-time TDDFT, as implemented in octopus, can be the method of choice for studying the excited states of large molecular systems in modern parallel architectures, XA and AA acknowledge hardware donations by Advanced Micro Devices, Inc. (AMD) and Nvidia Corp., and support from the US NSF CDI program (PHY-0835713). MALM acknowledges support from the French ANR (ANR-08-CEXC8-008-01) and computational resources from GENCI (project x2011096017). JAR acknowledges the scholarship of the University of the Basque Country UPV/EHU. DAS acknowledges support from the US NSF IGERT and Graduate Research Fellowship Programs, and by NSF Grant No. DMR 10-1006184. SGL was supported by NSF Grant No. DMR10-1006184 and by the Director, Office of Science, Office of Basic Energy Sciences, Materials Sciences and Engineering Division, US Department of Energy under Contract No. 10DE-AC02-05CH11231. The authors were partially supported by the European Research Council Advanced Grant DYNamo (ERC-2010-AdG—Proposal No. 267374), Spanish Grants (FIS2011-65702-C02-01 and PIB2010US-00652), ACI-Promociona (ACI2009-1036), Grupos Consolidados UPV/EHU del Gobierno Vasco (IT-319-07), University of the Basque Country UPV/EHU general funding for research groups ALDAPA (GIU10/02), Consolider nanoTHERM (Grant No. CSD2010-00044) and European Commission projects CRONOS (280879-2 CRONOS CP-FP7) and THEMA (FP7-NMP-2008-SMALL-2, 228539).

International audience; The implementation via CUDA of a hybrid dense dynamic programming method for knapsack problems on a multi-GPU architecture is considered. Tests are carried out on a Bull cluster with Tesla S1070 computing systems. A first series of computational results shows substantial speedup close to 30 with two GPUs.

The use of atomistic methods, such as the continuous cellular automaton (CCA), is currently regarded as an accurate and efficient approach for the simulation of anisotropic etching in the development of micro-electro-mechanical systems (MEMS). However, whenever the targeted etching condition is modified (e.g. by changing the substrate material, etchant type, concentration and/or temperature) this approach requires performing a time-consuming recalibration of the full set of internal atomistic rates defined within the method. Based on the level set (LS) approach as an alternative and using the experimental data directly as input, we present a fully operational simulator that exhibits similar accuracy to the latest CCA models. The proposed simulator is tested by describing a wide range of silicon and quartz MEMS structures obtained in different etchants through complex processes, including double-sided etching as well as different mask patterns during different etching steps and/or simultaneous masking materials on different regions of the substrate. The results demonstrate that the LS method is able to simulate anisotropic etching for complex MEMS processes with similar computational times and accuracy as the atomistic models. © 2013 IOP Publishing Ltd., This work has been supported by the Spanish FPI-MICINN BES-2011-045940 grant and the Ramón y Cajal Fellowship Program by the Spanish Ministry of Science and Innovation. Also, we acknowledge support by the JAE-Doc grant from the Junta para la Ampliación de Estudios program co-funded by FSE and the Professor Partnership Program by NVIDIA Corporation.

We combine experiments and simulations to study the acceleration of anisotropic etching of crystalline silicon at the mask-substrate interface, as a function of the coordination number of the substrate atoms located at the junction between obtuse-angled {1 1 1} facets and the mask layer. Atomistic simulations based on the use of the continuous cellular automaton (CCA) conclude that the interface atoms react faster with the etchant, thus initiating a step flow process that results in increased etch rates for the obtuse facets. By generating a wide range of complex cavities on high-index silicon wafers with a single-side, single-step etching, the comparison of the experimental and simulated results strongly indicates that the CCA method is suitable for accurately describing not only the development of micron-scaled structures but also, for the first time, the formation of submicron shapes. The study also describes the acceleration of obtuse facets formed through double-side etching, obtaining results in good agreement with previous experiments., This work was supported by the Ramón y Cajal Fellowship Program of the Spanish Ministry of Science and Innovation (M A Gosálvez), the Swiss National Science Foundation (Award Number 200021-146747), the JAE-Doc grant from the ’Junta para la Ampliación de Estudios’ program co-funded by FSE (N Ferrando), and the Professor Partnership Program by NVIDIA Corporation.

Manipulating textile objects with a robot is a challenging task, especially because the garment perception is difficult due to the endless configurations it can adopt, coupled with a large variety of colors and designs. Most current approaches follow a multiple re-grasp strategy, in which clothes are sequentially grasped from different points until one of them yields a recognizable configuration. In this work we propose a method that combines 3D and appearance information to directly select a suitable grasping point for the task at hand, which in our case consists of hanging a shirt or a polo shirt from a hook. Our method follows a coarse-to-fine approach in which, first, the collar of the garment is detected and, next, a grasping point on the lapel is chosen using a novel 3D descriptor. In contrast to current 3D descriptors, ours can run in real time, even when it needs to be densely computed over the input image. Our central idea is to take advantage of the structured nature of range images that most depth sensors provide and, by exploiting integral imaging, achieve speed-ups of two orders of magnitude with respect to competing approaches, while maintaining performance. This makes it especially adequate for robotic applications as we thoroughly demonstrate in the experimental section., This research is partially funded by the Spanish Ministry of Economy and Competitiveness under project TIN2014-58178-R, by the CSIC project MANIPlus (201350E102), and by the ERA-Net CHISTERA projects ViSen PCIN-2013-047 and I-DRESS PCIN-2015-147. A. Ramisa worked under the JAE-DOC grant from CSIC and FSE. The authors are grateful to the Nvidia donation program for its support with GPU cards.