Dissertationen zum Thema „Ensemble de données multimodal“
Geben Sie eine Quelle nach APA, MLA, Chicago, Harvard und anderen Zitierweisen an
Machen Sie sich mit Top-50 Dissertationen für die Forschung zum Thema "Ensemble de données multimodal" bekannt.
Neben jedem Werk im Literaturverzeichnis ist die Option "Zur Bibliographie hinzufügen" verfügbar. Nutzen Sie sie, wird Ihre bibliographische Angabe des gewählten Werkes nach der nötigen Zitierweise (APA, MLA, Harvard, Chicago, Vancouver usw.) automatisch gestaltet.
Sie können auch den vollen Text der wissenschaftlichen Publikation im PDF-Format herunterladen und eine Online-Annotation der Arbeit lesen, wenn die relevanten Parameter in den Metadaten verfügbar sind.
Sehen Sie die Dissertationen für verschiedene Spezialgebieten durch und erstellen Sie Ihre Bibliographie auf korrekte Weise.
Chen, Jianan. „Deep Learning Based Multimodal Retrieval“. Electronic Thesis or Diss., Rennes, INSA, 2023. http://www.theses.fr/2023ISAR0019.
Der volle Inhalt der QuelleMultimodal tasks play a crucial role in the progression towards achieving general artificial intelligence (AI). The primary goal of multimodal retrieval is to employ machine learning algorithms to extract relevant semantic information, bridging the gap between different modalities such as visual images, linguistic text, and other data sources. It is worth noting that the information entropy associated with heterogeneous data for the same high-level semantics varies significantly, posing a significant challenge for multimodal models. Deep learning-based multimodal network models provide an effective solution to tackle the difficulties arising from substantial differences in information entropy. These models exhibit impressive accuracy and stability in large-scale cross-modal information matching tasks, such as image-text retrieval. Furthermore, they demonstrate strong transfer learning capabilities, enabling a well-trained model from one multimodal task to be fine-tuned and applied to a new multimodal task, even in scenarios involving few-shot or zero-shot learning. In our research, we develop a novel generative multimodal multi-view database specifically designed for the multimodal referential segmentation task. Additionally, we establish a state-of-the-art (SOTA) benchmark and multi-view metric for referring expression segmentation models in the multimodal domain. The results of our comparative experiments are presented visually, providing clear and comprehensive insights
Wang, Xin. „Gaze based weakly supervised localization for image classification : application to visual recognition in a food dataset“. Thesis, Paris 6, 2017. http://www.theses.fr/2017PA066577/document.
Der volle Inhalt der QuelleIn this dissertation, we discuss how to use the human gaze data to improve the performance of the weak supervised learning model in image classification. The background of this topic is in the era of rapidly growing information technology. As a consequence, the data to analyze is also growing dramatically. Since the amount of data that can be annotated by the human cannot keep up with the amount of data itself, current well-developed supervised learning approaches may confront bottlenecks in the future. In this context, the use of weak annotations for high-performance learning methods is worthy of study. Specifically, we try to solve the problem from two aspects: One is to propose a more time-saving annotation, human eye-tracking gaze, as an alternative annotation with respect to the traditional time-consuming annotation, e.g. bounding box. The other is to integrate gaze annotation into a weakly supervised learning scheme for image classification. This scheme benefits from the gaze annotation for inferring the regions containing the target object. A useful property of our model is that it only exploits gaze for training, while the test phase is gaze free. This property further reduces the demand of annotations. The two isolated aspects are connected together in our models, which further achieve competitive experimental results
Costa, Daniel Moura Martins da. „Ensemble baseado em métodos de Kernel para reconhecimento biométrico multimodal“. Universidade de São Paulo, 2016. http://www.teses.usp.br/teses/disponiveis/100/100131/tde-28072016-190335/.
Der volle Inhalt der QuelleWith the advancement of technology, traditional strategies for identifying people become more susceptible to failure, in order to overcome these difficulties some approaches have been proposed in the literature. Among these approaches highlights the Biometrics. The field of Biometrics encompasses a wide variety of technologies used to identify and verify the person\'s identity through the measurement and analysis of physiological and behavioural aspects of the human body. As a result, biometrics has a wide field of applications in systems that require precise identification of their users. The most popular biometric systems are based on face recognition and fingerprint matching. Furthermore, there are other biometric systems that utilize iris and retinal scan, speech, face, and hand geometry. In recent years, biometrics authentication has seen improvements in reliability and accuracy, with some of the modalities offering good performance. However, even the best biometric modality is facing problems. Recently, big efforts have been undertaken aiming to employ multiple biometric modalities in order to make the authentication process less vulnerable to attacks. Multimodal biometrics is a relatively new approach to biometrics representation that consolidate multiple biometric modalities. Multimodality is based on the concept that the information obtained from different modalities complement each other. Consequently, an appropriate combination of such information can be more useful than using information from single modalities alone. The main issues involved in building a unimodal biometric System concern the definition of the feature extraction technique and type of classifier. In the case of a multimodal biometric System, in addition to these issues, it is necessary to define the level of fusion and fusion strategy to be adopted. The aim of this dissertation is to investigate the use of committee machines to fuse multiple biometric modalities, considering different fusion strategies, taking into account advanced methods in machine learning. In particular, it will give emphasis to the analyses of different types of machine learning methods based on Kernel and its organization into arrangements committee machines, aiming biometric authentication based on face, fingerprint and iris. The results showed that the proposed approach is capable of designing a multimodal biometric System with recognition rate than those obtained by the unimodal biometrics Systems.
Larsson, Sanna. „Kommunikation - vad är det? : En studie av musikalisk kommunikation i ensemblesammanhang“. Thesis, Karlstads universitet, Institutionen för konstnärliga studier, 2014. http://urn.kb.se/resolve?urn=urn:nbn:se:kau:diva-33627.
Der volle Inhalt der QuelleThe study focuses on how singers communicate with their fellow musicians in music ensembles considering the multimodal perspective of communication and learning conditions. With the help of video recordings of my own communication within an ensemble I have analysed which semiotic resources that are used and my underlying intentions with them. The result shows that the communication between the group members first are established after a while into the learning process and that the singing part shows to be the prime communicator for me in this case where dynamic, articulating and playfulness indicate security or insecurity about the form of the song and the melody. It also works as a guide to where I am at in the learning process. In the chapter on discussion I present, among other things, how the first ensemble lesson is created due to the multimodal perspective on favourable learning conditions and also the importance of body language for favourable communication in music ensembles.
Lösegård, Linus. „”Det är inte bara ord som används” : En studie av musiklärares sätt att kommunicera med sina elever i undervisning av ensemble“. Thesis, Karlstads universitet, Musikhögskolan Ingesund, 2013. http://urn.kb.se/resolve?urn=urn:nbn:se:kau:diva-28917.
Der volle Inhalt der QuelleThe study focuses on how teachers communicate when teaching music ensembles in which they meet several students simultaneously. With the help of recorded video lessons of four teachers while teaching small ensembles at the high school level, I have analysed the means of communication and how these are applied. The results show that teachers use verbal as well as nonverbal communication in their teaching. They communicate with their students during the lesson via several multimodal semiotic resources. The language of communication that results is a social, humorous, authoritative, creative, knowledgeable and vivid language. In the chapter on discussion, I present among other things, how the teachers in the study chose to design their education based on the situations regarding performance and through performance, and the conditions thus created for themselves and their students.
Aron, Michael. „Acquisition et modélisation de données articulatoires dans un contexte multimodal“. Phd thesis, Université Henri Poincaré - Nancy I, 2009. http://tel.archives-ouvertes.fr/tel-00432124.
Der volle Inhalt der QuelleAron, Michaël. „Acquisition et modélisation de données articulatoires dans un contexte multimodal“. Thesis, Nancy 1, 2009. http://www.theses.fr/2009NAN10097/document.
Der volle Inhalt der QuelleThere is no single technique that will allow all relevant behaviour of the speech articulators (lips, tongue, palate...) to be spatially ant temporally acquired. Thus, this thesis investigates the fusion of multimodal articulatory data. A framework is described in order to acquire and fuse automatically an important database of articulatory data. This includes: 2D Ultrasound (US) data to recover the dynamic of the tongue, stereovision data to recover the 3D dynamic of the lips, electromagnetic sensors that provide 3D position of points on the face and the tongue, and 3D Magnetic Resonance Imaging (MRI) that depict the vocal tract for various sustained articulations. We investigate the problems of the temporal synchronization and the spatial registration between all these modalities, and also the extraction of the shape articulators from the data (tongue tracking in US images). We evaluate the uncertainty of our system by quantifying the spatial and temporal inaccuracies of the components of the system, both individually and in combination. Finally, the fused data are evaluated on an existing articulatory model to assess their quality for an application in speech production
Bato, Mary Grace. „Vers une assimilation des données de déformation en volcanologie“. Thesis, Université Grenoble Alpes (ComUE), 2018. http://www.theses.fr/2018GREAU018/document.
Der volle Inhalt der QuelleTracking magma emplacement at shallow depth as well as its migration towards the Earth's surface is crucial to forecast volcanic eruptions.With the recent advances in Interferometric Synthetic Aperture Radar (InSAR) imaging and the increasing number of continuous Global Navigation Satellite System (GNSS) networks recorded on volcanoes, it is now possible to provide continuous and spatially extensive evolution of surface displacements during inter-eruptive periods. For basaltic volcanoes, these measurements combined with simple dynamical models can be exploited to characterise and to constrain magma pressure building within one or several magma reservoirs, allowing better predictive information on the emplacement of magma at shallow depths. Data assimilation—a sequential time-forward process that best combines models and observations, sometimes a priori information based on error statistics, to predict the state of a dynamical system—has recently gained popularity in various fields of geoscience (e.g. ocean-weather forecasting, geomagnetism and natural resources exploration). In this dissertation, I present the very first application of data assimilation in volcanology from synthetic tests to analyzing real geodetic data.The first part of this work focuses on the development of strategies in order to test the applicability and to assess the potential of data assimilation, in particular, the Ensemble Kalman Filter (EnKF) using a simple two-chamber dynamical model (Reverso2014) and artificial geodetic data. Synthetic tests are performed in order to address the following: 1) track the magma pressure evolution at depth and reconstruct the synthetic ground surface displacements as well as estimate non-evolving uncertain model parameters, 2) properly assimilate GNSS and InSAR data, 3) highlight the strengths and weaknesses of EnKF in comparison with a Bayesian-based inversion technique (e.g. Markov Chain Monte Carlo). Results show that EnKF works well with the synthetic cases and there is a great potential in utilising data assimilation for real-time monitoring of volcanic unrest.The second part is focused on applying the strategy that we developed through synthetic tests in order to forecast the rupture of a magma chamber in real time. We basically explored the 2004-2011 inter-eruptive dataset at Grímsvötn volcano in Iceland. Here, we introduced the concept of “eruption zones” based on the evaluation of the probability of eruption at each time step estimated as the percentage of model ensembles that exceeded their failure overpressure values initially assigned following a given distribution. Our results show that when 25 +/- 1% of the model ensembles exceeded the failure overpressure, an actual eruption is imminent. Furthermore, in this chapter, we also extend the previous synthetic tests by further enhancing the EnKF strategy of assimilating geodetic data in order to adapt to real world problems such as, the limited amount of geodetic data available to monitor ice-covered active volcanoes. Common diagnostic tools in data assimilation are presented.Finally, I demonstrate that in addition to the interest of predicting volcanic eruptions, sequential assimilation of geodetic data on the basis of EnKF shows a unique potential to give insights into volcanic system roots. Using the two-reservoir dynamical model for Grímsvötn 's plumbing system and assuming a fixed geometry and constant magma properties, we retrieve the temporal evolution of the basal magma inflow beneath Grímsvötn that drops up to 85% during the 10 months preceding the initiation of the Bárdarbunga rifting event. The loss of at least 0.016 km3 in the magma supply of Grímsvötn is interpreted as a consequence of magma accumulation beneath Bárdarbunga and subsequent feeding of the Holuhraun eruption 41 km away
Pinson, Franck. „Ajustement de primitives d'objets de forme libre sur un ensemble de données réelles“. Compiègne, 1989. http://www.theses.fr/1989COMPD179.
Der volle Inhalt der QuelleLê, Thanh Vu. „Visualisation interactive 3D pour un ensemble de données géographiques de très grande taille“. Pau, 2011. http://www.theses.fr/2011PAUU3005.
Der volle Inhalt der QuelleReal-time terrain rendering remains an active area of research for a lot of modern computer based applications such as geographic information systems (GIS), interactive 3D games, flights simulators or virtual reality. The technological breakthroughs in data aquisition, coupled with recent advances in display technology have simultaneously led to substantial increases in resolution of both the Digital Elevation Models (DEM) and the various displays used to present this information. In this phD, we have presented a new out-of-core terrain visualization algorithm that achieves per-pixel accurate shading of large textured elevation maps in real-time : our first contribution is the LOD scheme which is based on a small precomputed quadtree of geometric errors, whose nodes are selected for asynchronous loading and rendering depending on a projection in screenspace of those errors. The terrain data and its color texture are manipulated by the CPU in a unified manner as a collection of raster image patches, whose dimensions depends on their screen-space occupancy ; our second contribution is a novel method to remove artifacts that appear on the border between quadtree blocks, we generate a continuous surface without needing additional mesh ; our latest contribution is an effective method adapted to our data structure for the geomorphing, it can be implemented entirely on the GPU. The presented framework exhibits several interesting features over other existing techniques : there is no mesh manipulation or mesh data structures required ; terrain geometric complexity only depends on projected elevation error views from above result in very coarse meshes), lower geometric complexity degrades terrain silhouettes but not details brought in through normal map shading, real-time rendering with support for progressive data loading ; and geometric information and color textures are similarly and efficiently handled as raster data by the CPU. Due to simplified data structures, the system is compact, CPU and GPU efficient and is simple to implement
Drouet, d'Aubigny Gérard. „L'analyse multidimensionnelle des données de dissimilarité : [thèse soutenue sur un ensemble de travaux]“. Grenoble 1, 1989. http://tel.archives-ouvertes.fr/tel-00332393.
Der volle Inhalt der QuelleRabah, Mazouzi. „Approches collaboratives pour la classification des données complexes“. Thesis, Paris 8, 2016. http://www.theses.fr/2016PA080079.
Der volle Inhalt der QuelleThis thesis focuses on the collaborative classification in the context of complex data, in particular the context of Big Data, we used some computational paradigms to propose new approaches based on HPC technologies. In this context, we aim at offering massive classifiers in the sense that the number of elementary classifiers that make up the multiple classifiers system can be very high. In this case, conventional methods of interaction between classifiers is no longer valid and we had to propose new forms of interaction, where it is not constrain to take all classifiers predictions to build an overall prediction. According to this, we found ourselves faced with two problems: the first is the potential of our approaches to scale up. The second, is the diversity that must be created and maintained within the system, to ensure its performance. Therefore, we studied the distribution of classifiers in a cloud-computing environment, this multiple classifiers system can be massive and their properties are those of a complex system. In terms of diversity of data, we proposed a training data enrichment approach for the generation of synthetic data from analytical models that describe a part of the phenomenon studied. so, the mixture of data reinforces learning classifiers. The experimentation made have shown the great potential for the substantial improvement of classification results
Barbe, Philippe. „Ensemble d'information de marché et détermination des taux de change : l'apport des données d'enquête“. Bordeaux 4, 1997. http://www.theses.fr/1997BOR40011.
Der volle Inhalt der QuelleThe object of this thesis is to study the functionning of the foreign exchange market when agents don't know the "true model". So we reject the rational expectations hypothesis for exchange rate. Therefore, it is important to know how agents on the market form their information set. To answer to this question, we adopt a methodology based on survey data, and a concept of restricted rationality, the economically rational expectations. Our results are the following. First of all, the market information set has two components, a fundamental component and a technical component. The first component is based on economic indicators analysis and command the anticipation function when data are available on the market. The second component is important when fondamental information are not available. Furthermore, the analysis of market expectations shows that these variables are self-fulfilling in short-term and more heterogeneous in long-term
Farchi, Alban. „On the localisation of ensemble data assimilation methods“. Thesis, Paris Est, 2019. http://www.theses.fr/2019PESC1034.
Der volle Inhalt der QuelleData assimilation is the mathematical discipline which gathers all the methods designed to improve the knowledge of the state of a dynamical system using both observations and modelling results of this system. In the geosciences, data assimilation it mainly applied to numerical weather prediction. It has been used in operational centres for several decades, and it has significantly contributed to the increase in quality of the forecasts.Ensemble methods are powerful tools to reduce the dimension of the data assimilation systems. Currently, the two most widespread classes of ensemble data assimilation methods are the ensemble Kalman filter (EnKF) and the particle filter (PF). The success of the EnKF in high-dimensional geophysical systems is largely due to the use of localisation. Localisation is based on the assumption that correlations between state variables in a dynamical system decrease at a fast rate with the distance. In this thesis, we have studied and improved localisation methods for ensemble data assimilation.The first part is dedicated to the implementation of localisation in the PF. The recent developments in local particle filtering are reviewed, and a generic and theoretical classification of local PF algorithms is introduced, with an emphasis on the advantages and drawbacks of each category. Alongside the classification, practical solutions to the difficulties of local particle filtering are suggested. The local PF algorithms are tested and compared using twin experiments with low- to medium-order systems. Finally, we consider the case study of the prediction of the tropospheric ozone using concentration measurements. Several data assimilation algorithms, including local PF algorithms, are applied to this problem and their performances are compared.The second part is dedicated to the implementation of covariance localisation in the EnKF. We show how covariance localisation can be efficiently implemented in the deterministic EnKF using an augmented ensemble. The proposed algorithm is tested using twin experiments with a medium-order model and satellite-like observations. Finally, the consistency of the deterministic EnKF with covariance localisation is studied in details. A new implementation is proposed and compared to the original one using twin experiments with low-order models
Pomorski, Denis. „Apprentissage automatique symbolique/numérique : construction et évaluation d'un ensemble de règles à partir des données“. Lille 1, 1991. http://www.theses.fr/1991LIL10117.
Der volle Inhalt der QuelleCastanié, Laurent. „Visualisation de données volumiques massives : application aux données sismiques“. Thesis, Vandoeuvre-les-Nancy, INPL, 2006. http://www.theses.fr/2006INPL083N/document.
Der volle Inhalt der QuelleSeismic reflection data are a valuable source of information for the three-dimensional modeling of subsurface structures in the exploration-production of hydrocarbons. This work focuses on the implementation of visualization techniques for their interpretation. We face both qualitative and quantitative challenges. It is indeed necessary to consider (1) the particular nature of seismic data and the interpretation process (2) the size of data. Our work focuses on these two distinct aspects : 1) From the qualitative point of view, we first highlight the main characteristics of seismic data. Based on this analysis, we implement a volume visualization technique adapted to the specificity of the data. We then focus on the multimodal aspect of interpretation which consists in combining several sources of information (seismic and structural). Depending on the nature of these sources (strictly volumes or both volumes and surfaces), we propose two different visualization systems. 2) From the quantitative point of view, we first define the main hardware constraints involved in seismic interpretation. Focused on these constraints, we implement a generic memory management system. Initially able to couple visualization and data processing on massive data volumes, it is then improved and specialised to build a dynamic system for distributed memory management on PC clusters. This later version, dedicated to visualization, allows to manipulate regional scale seismic data (100-200 GB) in real-time. The main aspects of this work are both studied in the scientific context of visualization and in the application context of geosciences and seismic interpretation
Faucon, Jean-Christophe. „Etudes statistiques et des relations structure-écotoxicité appliquées aux données écotoxicologiques d'un ensemble hétérogène de substances nouvelles“. Caen, 1998. http://www.theses.fr/1998CAEN4002.
Der volle Inhalt der QuelleAsgari, Fereshteh. „Inferring user multimodal trajectories from cellular network metadata in metropolitan areas“. Thesis, Evry, Institut national des télécommunications, 2016. http://www.theses.fr/2016TELE0005/document.
Der volle Inhalt der QuelleAround half of the world population is living in cities where different transportation networks are cooperating together to provide some efficient transportation facilities for individuals. To improve the performance of the multimodal transportation network it is crucial to monitor and analyze the multimodal trajectories, however obtaining the multimodal mobility data is not a trivial task. GPS data with fine accuracy, is extremely expensive to collect; Additionally, GPS is not available in tunnels and underground. Recently, thanks to telecommunication advancement cellular dataset such as Call Data Records (CDRs), is a great resource of mobility data, nevertheless, this kind of dataset is noisy and sparse in time. Our objective in this thesis is to propose a solution to this challenging issue of inferring real trajectory and transportation layer from wholly cellular observation. To achieve these objectives, we use Cellular signalization data which is more frequent than CDRs and despite their spatial inaccuracy, they provide a fair source of multimodal trajectory data. We propose 'CT-Mapper’ to map cellular signalization data collected from smart phones over the multimodal transportation network. Our proposed algorithm uses Hidden Markov Model property and topological properties of different transportation layers to model an unsupervised mapping algorithm which maps sparse cellular trajectories on multilayer transportation network. Later on, we propose ‘LCT-Mapper’ an algorithm to infer the main mode of trajectories. The area of study in this research work is Paris and region (Ile-de-France); we have modeled and built the multimodal transportation network database. To evaluate our proposed algorithm, we use real trajectories data sets collected from a group of volunteers for a period of 1 month. The user's cellular signalization data was provided by a french operator to assess the performance of our proposed algorithms using GPS data as ground truth. An extensive set of evaluation has been performed to validate the proposed algorithms. To summarize, we have shown in this work that it is feasible to infer the multimodal trajectory of users in an unsupervised manner. Our achievement makes it possible to investigate the multimodal mobility behavior of people and explore and monitor the population flow over multilayer transportation network
Raharjo, Agus Budi. „Reliability in ensemble learning and learning from crowds“. Electronic Thesis or Diss., Aix-Marseille, 2019. http://www.theses.fr/2019AIXM0606.
Der volle Inhalt der QuelleThe combination of several human expert labels is generally used to make reliable decisions. However, using humans or learning systems to improve the overall decision is a crucial problem. Indeed, several human experts or machine learning have not necessarily the same performance. Hence, a great effort is made to deal with this performance problem in the presence of several actors, i.e., humans or classifiers. In this thesis, we present the combination of reliable classifiers in ensemble learning and learning from crowds. The first contribution is a method, based on weighted voting, which allows selecting a reliable combination of classifications. Our algorithm RelMV transforms confidence scores, obtained during the training phase, into reliable scores. By using these scores, it determines a set of reliable candidates through both static and dynamic selection process. When it is hard to find expert labels as ground truth, we propose an approach based on Bayesian and expectation-maximization (EM) as our second contribution. The aim is to evaluate the reliability degree of each annotator and to aggregate the appropriate labels carefully. We optimize the computation time of the algorithm in order to adapt a large number of data collected from crowds. The obtained outcomes show better accuracy, stability, and computation time compared to the previous methods. Also, we conduct an experiment considering the melanoma diagnosis problem using a real-world medical dataset consisting of a set of skin lesions images, which is annotated by multiple dermatologists
Nadal, Robert. „Analyse des données astronomiques contenues dans le "Commentaire" d'Hipparque : [thèse en partie soutenue sur un ensemble de travaux]“. Toulouse 3, 1990. http://www.theses.fr/1990TOU30197.
Der volle Inhalt der QuelleAfli, Haithem. „La Traduction automatique statistique dans un contexte multimodal“. Thesis, Le Mans, 2014. http://www.theses.fr/2014LEMA1012/document.
Der volle Inhalt der QuelleThe performance of Statistical Machine Translation Systems statistics depends on the availability of bilingual parallel texts, also known as bitexts. However, freely available parallel texts are also a sparse resource : the size is often limited, languistic coverage insufficient or the domain of texts is not appropriate. There are relatively few pairs of languages for which parallel corpora sizes are available for some domains. One way to overcome the lack of parallel data is to exploit comparable corpus that are more abundant. Previous work in this area have been applied for the text modality. The question we asked in this thesis is : can comparable multimodal corpus allows us to make solutions to the lack of parallel data in machine translation? In this thesis, we studied how to use resources from different modalities (text or speech) for the development of a Statistical machine translation System. The first part of the contributions is to provide a method for extracting parallel data from a comparable multimodal corpus (text and audio). The audio data are transcribed with an automatic speech recognition system and translated with a machine translation system. These translations are then used as queries to select parallel sentences and generate a bitext. In the second part of the contribution, we aim to improve our method to exploit the sub-sentential entities creating an extension of our system to generate parallel segments. We also improve the filtering module. Finally, we présent several approaches to adapt translation systems with the extracted data. Our experiments were conducted on data from the TED and Euronews web sites which show the feasibility of our approaches
Feng, Wei. „Investigation of training data issues in ensemble classification based on margin concept : application to land cover mapping“. Thesis, Bordeaux 3, 2017. http://www.theses.fr/2017BOR30016/document.
Der volle Inhalt der QuelleClassification has been widely studied in machine learning. Ensemble methods, which build a classification model by integrating multiple component learners, achieve higher performances than a single classifier. The classification accuracy of an ensemble is directly influenced by the quality of the training data used. However, real-world data often suffers from class noise and class imbalance problems. Ensemble margin is a key concept in ensemble learning. It has been applied to both the theoretical analysis and the design of machine learning algorithms. Several studies have shown that the generalization performance of an ensemble classifier is related to the distribution of its margins on the training examples. This work focuses on exploiting the margin concept to improve the quality of the training set and therefore to increase the classification accuracy of noise sensitive classifiers, and to design effective ensemble classifiers that can handle imbalanced datasets. A novel ensemble margin definition is proposed. It is an unsupervised version of a popular ensemble margin. Indeed, it does not involve the class labels. Mislabeled training data is a challenge to face in order to build a robust classifier whether it is an ensemble or not. To handle the mislabeling problem, we propose an ensemble margin-based class noise identification and elimination method based on an existing margin-based class noise ordering. This method can achieve a high mislabeled instance detection rate while keeping the false detection rate as low as possible. It relies on the margin values of misclassified data, considering four different ensemble margins, including the novel proposed margin. This method is extended to tackle the class noise correction which is a more challenging issue. The instances with low margins are more important than safe samples, which have high margins, for building a reliable classifier. A novel bagging algorithm based on a data importance evaluation function relying again on the ensemble margin is proposed to deal with the class imbalance problem. In our algorithm, the emphasis is placed on the lowest margin samples. This method is evaluated using again four different ensemble margins in addressing the imbalance problem especially on multi-class imbalanced data. In remote sensing, where training data are typically ground-based, mislabeled training data is inevitable. Imbalanced training data is another problem frequently encountered in remote sensing. Both proposed ensemble methods involving the best margin definition for handling these two major training data issues are applied to the mapping of land covers
Chlaily, Saloua. „Modèle d'interaction et performances du traitement du signal multimodal“. Thesis, Université Grenoble Alpes (ComUE), 2018. http://www.theses.fr/2018GREAT026/document.
Der volle Inhalt der QuelleThe joint processing of multimodal measurements is supposed to lead to better performances than those obtained using a single modality or several modalities independently. However, in literature, there are examples that show that is not always true. In this thesis, we analyze, in terms of mutual information and estimation error, the different situations of multimodal analysis in order to determine the conditions to achieve the optimal performances.In the first part, we consider the simple case of two or three modalities, each associated with noisy measurement of a signal. These modalities are linked through the correlations between the useful parts of the signal and the correlations between the noises. We show that the performances are improved if the links between the modalities are exploited. In the second part, we study the impact on performance of wrong links between modalities. We show that these false assumptions decline the performance, which can become lower than the performance achieved using a single modality.In the general case, we model the multiple modalities as a noisy Gaussian channel. We then extend literature results by considering the impact of the errors on signal and noise probability densities on the information transmitted by the channel. We then analyze this relationship in the case of a simple model of two modalities. Our results show in particular the unexpected fact that a double mismatch of the noise and the signal can sometimes compensate for each other, and thus lead to very good performances
Guillaumin, Matthieu. „Données multimodales pour l'analyse d'image“. Phd thesis, Grenoble, 2010. http://tel.archives-ouvertes.fr/tel-00522278/en/.
Der volle Inhalt der QuelleGuillaumin, Matthieu. „Données multimodales pour l'analyse d'image“. Phd thesis, Grenoble, 2010. http://www.theses.fr/2010GRENM048.
Der volle Inhalt der QuelleThis dissertation delves into the use of textual metadata for image understanding. We seek to exploit this additional textual information as weak supervision to improve the learning of recognition models. There is a recent and growing interest for methods that exploit such data because they can potentially alleviate the need for manual annotation, which is a costly and time-consuming process. We focus on two types of visual data with associated textual information. First, we exploit news images that come with descriptive captions to address several face related tasks, including face verification, which is the task of deciding whether two images depict the same individual, and face naming, the problem of associating faces in a data set to their correct names. Second, we consider data consisting of images with user tags. We explore models for automatically predicting tags for new images, i. E. Image auto-annotation, which can also used for keyword-based image search. We also study a multimodal semi-supervised learning scenario for image categorisation. In this setting, the tags are assumed to be present in both labelled and unlabelled training data, while they are absent from the test data. Our work builds on the observation that most of these tasks can be solved if perfectly adequate similarity measures are used. We therefore introduce novel approaches that involve metric learning, nearest neighbour models and graph-based methods to learn, from the visual and textual data, task-specific similarities. For faces, our similarities focus on the identities of the individuals while, for images, they address more general semantic visual concepts. Experimentally, our approaches achieve state-of-the-art results on several standard and challenging data sets. On both types of data, we clearly show that learning using additional textual information improves the performance of visual recognition systems
Létourneau, François. „Analyse du potentiel de l'approche entrepôt de données pour l'intégration des métadonnées provenant d'un ensemble de géorépertoires disponibles sur Internet“. Thesis, National Library of Canada = Bibliothèque nationale du Canada, 1998. http://www.collectionscanada.ca/obj/s4/f2/dsk2/tape17/PQDD_0007/MQ31752.pdf.
Der volle Inhalt der QuelleLe, Brun Alexia. „Etude d'un ensemble de paramètres liés à la sécheresse de la peau : traitement des données par des méthodes d'analyses multidimensionnelles“. Bordeaux 1, 1986. http://www.theses.fr/1986BOR10880.
Der volle Inhalt der QuelleLEBRUN, ALEXIA MARIE. „Etude d'un ensemble de paramètres liés a la sécheresse de la peau : traitements des données par des méthodes d'analyses multidimensionnelles“. Bordeaux 1, 1986. http://www.theses.fr/1986BOR10885.
Der volle Inhalt der QuelleLe, Brun Alexia. „Étude d'un ensemble de paramètres liés à la sécheresse de la peau : traitements des données par des méthodes d'analyses multidimensionnelles“. Bordeaux 1, 1986. http://www.theses.fr/1986BOR10689.
Der volle Inhalt der QuelleBergou, El Houcine. „Méthodes numériques pour les problèmes des moindres carrés, avec application à l'assimilation de données“. Thesis, Toulouse, INPT, 2014. http://www.theses.fr/2014INPT0114/document.
Der volle Inhalt der QuelleThe Levenberg-Marquardt algorithm (LM) is one of the most popular algorithms for the solution of nonlinear least squares problems. Motivated by the problem structure in data assimilation, we consider in this thesis the extension of the LM algorithm to the scenarios where the linearized least squares subproblems, of the form min||Ax - b ||^2, are solved inexactly and/or the gradient model is noisy and accurate only within a certain probability. Under appropriate assumptions, we show that the modified algorithm converges globally and almost surely to a first order stationary point. Our approach is applied to an instance in variational data assimilation where stochastic models of the gradient are computed by the so-called ensemble Kalman smoother (EnKS). A convergence proof in L^p of EnKS in the limit for large ensembles to the Kalman smoother is given. We also show the convergence of LM-EnKS approach, which is a variant of the LM algorithm with EnKS as a linear solver, to the classical LM algorithm where the linearized subproblem is solved exactly. The sensitivity of the trucated sigular value decomposition method to solve the linearized subprobems is studied. We formulate an explicit expression for the condition number of the truncated least squares solution. This expression is given in terms of the singular values of A and the Fourier coefficients of b
Séraphin, John. „Réalisation d'un intranet : cohérence d'un ensemble réparti et communicant, autour d'une architecture réflexive“. Paris 5, 1998. http://www.theses.fr/1998PA05S007.
Der volle Inhalt der QuelleVielzeuf, Valentin. „Apprentissage neuronal profond pour l'analyse de contenus multimodaux et temporels“. Thesis, Normandie, 2019. http://www.theses.fr/2019NORMC229/document.
Der volle Inhalt der QuelleOur perception is by nature multimodal, i.e. it appeals to many of our senses. To solve certain tasks, it is therefore relevant to use different modalities, such as sound or image.This thesis focuses on this notion in the context of deep learning. For this, it seeks to answer a particular problem: how to merge the different modalities within a deep neural network?We first propose to study a problem of concrete application: the automatic recognition of emotion in audio-visual contents.This leads us to different considerations concerning the modeling of emotions and more particularly of facial expressions. We thus propose an analysis of representations of facial expression learned by a deep neural network.In addition, we observe that each multimodal problem appears to require the use of a different merge strategy.This is why we propose and validate two methods to automatically obtain an efficient fusion neural architecture for a given multimodal problem, the first one being based on a central fusion network and aimed at preserving an easy interpretation of the adopted fusion strategy. While the second adapts a method of neural architecture search in the case of multimodal fusion, exploring a greater number of strategies and therefore achieving better performance.Finally, we are interested in a multimodal view of knowledge transfer. Indeed, we detail a non-traditional method to transfer knowledge from several sources, i.e. from several pre-trained models. For that, a more general neural representation is obtained from a single model, which brings together the knowledge contained in the pre-trained models and leads to state-of-the-art performances on a variety of facial analysis tasks
Boubaker, Aimen. „Modelisation des composants mono-electroniques : Single-Electron Transistor et Single-Electron Memory“. Lyon, INSA, 2010. http://theses.insa-lyon.fr/publication/2010ISAL0046/these.pdf.
Der volle Inhalt der Quelle[This work concerns the study of SET/SEM single electron memories for CMOS technologies. The first part presents a review of quantum and Coulomb blockade effects in electronic nanodevices. In a second part, we present the main electrical models proposed for single electron devices. A comparison between semiconductor-based and metall ic-based single electron transistors. The third part of the thesis presents the SET/SEM memory structure on the basis of SIMON simulations. The device consists on the coupling of a metallic SET operating at high temperature with a metalli c memory node. Finnaly, an optimized memory device has been proposed in the Ti/Tiüx system. The proposed memory is able to write and erase a discrete number of electrons varying from 0 to 7 at room temperature. This opens the possibility of multilevel memory circuits. Finally, we have studied the data retenti on performances of the memory in the last part of this thesis. After the first simulations with the Ti/Tiüx materials system, we have simulated various metallic systems such as Pt, Au, TiSi2, and NiSi. We have shown that finally, the Ti/Ti02 systems gives the best data retention performances even at high temperatures, up to 430K. . ]
Balakrishnan, Arjun. „Integrity Analysis of Data Sources in Multimodal Localization System“. Thesis, université Paris-Saclay, 2020. http://www.theses.fr/2020UPASG060.
Der volle Inhalt der QuelleIntelligent vehicles are a key component in humanity’s vision for safer, efficient, and accessible transportation systems across the world. Due to the multitude of data sources and processes associated with Intelligent vehicles, the reliability of the total system is greatly dependent on the possibility of errors or poor performances observed in its components. In our work, we focus on the critical task of localization of intelligent vehicles and address the challenges in monitoring the integrity of data sources used in localization. The primary contribution of our research is the proposition of a novel protocol for integrity by combining integrity concepts from information systems with the existing integrity concepts in the field of Intelligent Transport Systems (ITS). An integrity monitoring framework based on the theorized integrity protocol that can handle multimodal localization problems is formalized. As the first step, a proof of concept for this framework is developed based on cross-consistency estimation of data sources using polynomial models. Based on the observations from the first step, a 'Feature Grid' data representation is proposed in the second step and a generalized prototype for the framework is implemented. The framework is tested in highways as well as complex urban scenarios to demonstrate that the proposed framework is capable of providing continuous integrity estimates of multimodal data sources used in intelligent vehicle localization
Samper, González Jorge Alberto. „Learning from multimodal data for classification and prediction of Alzheimer's disease“. Electronic Thesis or Diss., Sorbonne université, 2019. http://www.theses.fr/2019SORUS361.
Der volle Inhalt der QuelleAlzheimer's disease (AD) is the first cause of dementia worldwide, affecting over 20 million people. Its diagnosis at an early stage is essential to ensure a proper care of patients, and to develop and test novel treatments. AD is a complex disease that has to be characterized by the use of different measurements: cognitive and clinical tests, neuroimaging including magnetic resonance imaging (MRI) and positron emission tomography (PET), genotyping, etc. There is an interest in exploring the discriminative and predictive capabilities of these diverse markers, which reflect different aspects of the disease and potentially carry complementary information, from an early stage of the disease. The objective of this PhD thesis was thus to assess the potential and to integrate multiple modalities using machine learning methods, in order to automatically classify patients with AD and predict the development of the disease from the earliest stages. More specifically, we aimed to make progress toward the translation of such approaches toward clinical practice. The thesis comprises three main studies. The first one tackles the differential diagnosis between different forms of dementia from MRI data. This study was performed using clinical routine data, thereby providing a more realistic evaluation scenario. The second one proposes a new framework for reproducible evaluation of AD classification algorithms from MRI and PET data. Indeed, while numerous approaches have been proposed for AD classification in the literature, they are difficult to compare and to reproduce. The third part is devoted to the prediction of progression to AD in patients with mild cognitive impairment through the integration of multimodal data, including MRI, PET, clinical/cognitive evaluations and genotyping. In particular, we systematically assessed the added value of neuroimaging over clinical/cognitive data only. Since neuroimaging is more expensive and less widely available, this is important to justify its use as input of classification algorithms
Yang, Yin. „Study of variational ensemble methods for image assimilation“. Thesis, Rennes 1, 2014. http://www.theses.fr/2014REN1S118/document.
Der volle Inhalt der QuelleThe hybrid methods combing the 4D variational method and the ensemble Kalman filter provide a flexible framework. In such framework the potential advantages with respect to each method (e.g. the flow-dependent background error covariance, the ability to explicitly get the analysis error covariance matrix, the iterative minimization procedure and the simultaneously assimilation of all observations with in a time span etc.) can be retained. In this thesis, an enhanced ensemble-based 4DVar scheme is proposed and has been analyzed in detail in the case of the 2D shallow water model. Several variations related to this method are introduced and tested. We proposed a new nested loop scheme in which the background error covariance matrix is updated for each outer loop. We also devised different ensemble update schemes together with two localization schemes. And we exploited the links between the analysis error covariance matrix and the inverse Hessian of our 4D cost function. All these variants have been tested with the real Kinect-captured image data and synthetic image data associated with a SQG (Surface Quasi-Geostrophic) model, respectively. At the second stage, a parameter estimation scheme of our proposed ensemble-based variational method is devised. Such formulation allows the parameter estimation of an uncertainty subgrid stress tensor. And this uncertainty subgrid stress tensor is derived from a perspective of flow phenomenon driven by a stochastic process. Finally, a first effort is made to assimilation high-resolution image data with the dynamical model running on a much coarser grid
Oger, Niels. „Observation adaptative : limites de la prévision et du contrôle des incertitudes“. Phd thesis, Toulouse, INPT, 2015. http://oatao.univ-toulouse.fr/14491/1/oger.pdf.
Der volle Inhalt der QuelleMarty, Jean-Charles. „Evolution thermique de la lithosphère océanique : analyse des données altimétriques SEASAT et des donnees topographiques : [thèse en partie soutenue sur un ensemble de travaux]“. Toulouse 3, 1990. http://www.theses.fr/1990TOU30005.
Der volle Inhalt der QuelleAnzid, Hanan. „Fusion de données multimodales par combinaison de l’incertain et de modèles de perception“. Thesis, Bourgogne Franche-Comté, 2019. http://www.theses.fr/2019UBFCK046.
Der volle Inhalt der QuelleThe general idea is to use together heterogeneous multiple information on the same problem tainted by imperfections and coming from several sources in order to improve the knowledge of a given situation. Appropriate visualization of the images to aid in decision making using the perceptual information carried by the salience maps
Al-Najdi, Atheer. „Une approche basée sur les motifs fermés pour résoudre le problème de clustering par consensus“. Thesis, Université Côte d'Azur (ComUE), 2016. http://www.theses.fr/2016AZUR4111/document.
Der volle Inhalt der QuelleClustering is the process of partitioning a dataset into groups, so that the instances in the same group are more similar to each other than to instances in any other group. Many clustering algorithms were proposed, but none of them proved to provide good quality partition in all situations. Consensus clustering aims to enhance the clustering process by combining different partitions obtained from different algorithms to yield a better quality consensus solution. In this work, a new consensus clustering method, called MultiCons, is proposed. It uses the frequent closed itemset mining technique in order to discover the similarities between the different base clustering solutions. The identified similarities are presented in a form of clustering patterns, that each defines the agreement between a set of base clusters in grouping a set of instances. By dividing these patterns into groups based on the number of base clusters that define the pattern, MultiCons generates a consensussolution from each group, resulting in having multiple consensus candidates. These different solutions are presented in a tree-like structure, called ConsTree, that facilitates understanding the process of building the multiple consensuses, and also the relationships between the data instances and their structuring in the data space. Five consensus functions are proposed in this work in order to build a consensus solution from the clustering patterns. Approach 1 is to just merge any intersecting clustering patterns. Approach 2 can either merge or split intersecting patterns based on a proposed measure, called intersection ratio
Guillaume, Cécile. „Liens entre émotions et mémoire chez le jeune adulte et dans le vieillissement normal : données comportementales et électrophysiologiques : [thèse soutenue sur un ensemble de travaux]“. Caen, 2009. http://www.theses.fr/2009CAEN1537.
Der volle Inhalt der QuelleFourty, Thierry. „Estimation du contenu biochimique d'un couvert végétal à partir de données haute résolution spectrale acquises au niveau satellitaire : [thèse soutenue sur un ensemble de travaux]“. Toulouse 3, 1996. http://www.theses.fr/1996TOU30213.
Der volle Inhalt der QuelleMasri, Ali. „Multi-Network integration for an Intelligent Mobility“. Thesis, Université Paris-Saclay (ComUE), 2017. http://www.theses.fr/2017SACLV091/document.
Der volle Inhalt der QuelleMultimodality requires the integration of heterogeneous transportation data and services to construct a broad view of the transportation network. Many new transportation services (e.g. ridesharing, car-sharing, bike-sharing) are emerging and gaining a lot of popularity since in some cases they provide better trip solutions.However, these services are still isolated from the existing multimodal solutions and are proposed as alternative plans without being really integrated in the suggested plans. The concept of open data is raising and being adopted by many companies where they publish their data sources to the web in order to gain visibility. The goal of this thesis is to use these data to enable multimodality by constructing an extended transportation network that links these new services to existing ones.The challenges we face mainly arise from the integration problem in both transportation services and transportation data
Bonan, Bertrand. „Assimilation de données pour l'initialisation et l'estimation de paramètres d'un modèle d'évolution de calotte polaire“. Phd thesis, Université de Grenoble, 2013. http://tel.archives-ouvertes.fr/tel-00930097.
Der volle Inhalt der QuelleBendriss, Sabri. „Contribution à l'analyse et la conception d'un système d'information pour la gestion de la traçabilité des marchandises dans un contexte de transport multimodal“. Le Havre, 2009. http://www.theses.fr/2009LEHA0024.
Der volle Inhalt der QuelleOne of solutions to regulate and rationalize the physical displacement of goods is to succeed to synchronize the physical flow with its informational counterpart throughout the various links constituting the transport chain. In this context, a solution of goods tracking and tracing can contribute to a better mastery of flows. In this memory, we propose a data modeling approach for goods traceability based on innovative research approaches (PLM, Intelligent product, Product centered systems) and taking into account the possibilities offered by the use of NICT in terms of data sharing, auto-identification and geolocation. Then, in order to integrate our traceability data with the other transport chain data, but also in order to facilitate the sharing and the exchange of our data, we propose a modeling and the development of an intermediation platform based on the web services logic. Meeting criteria of interoperability and integrability, the result allows through mechanisms for exchange and saving data to follow and to restore goods lifecycle in its entirety
Benkaraache, Taoufik. „Problèmes de validité en classification hiérarchique et quelques généralisations aux ultramétriques à valeurs dans un ensemble ordonné“. Grenoble 1, 1993. http://www.theses.fr/1993GRE10102.
Der volle Inhalt der QuelleHaussaire, Jean-Matthieu. „Méthodes variationnelles d'ensemble itératives pour l'assimilation de données non-linéaire : Application au transport et la chimie atmosphérique“. Thesis, Paris Est, 2017. http://www.theses.fr/2017PESC1097/document.
Der volle Inhalt der QuelleData assimilation methods are constantly evolving to adapt to the various application domains. In atmospheric sciences, each new algorithm has first been implemented on numerical weather prediction models before being ported to atmospheric chemistry models. It has been the case for 4D variational methods and ensemble Kalman filters for instance. The new 4D ensemble variational methods (4D EnVar) are no exception. They were developed to take advantage of both variational and ensemble approaches and they are starting to be used in operational weather prediction centers, but have yet to be tested on operational atmospheric chemistry models.The validation of new data assimilation methods on these models is indeed difficult because of the complexity of such models. It is hence necessary to have at our disposal low-order models capable of synthetically reproducing key physical phenomenons from operational models while limiting some of their hardships. Such a model, called L95-GRS, has therefore been developed. It combines the simple meteorology from the Lorenz-95 model to a tropospheric ozone chemistry module with 7 chemical species. Even though it is of low dimension, it reproduces some of the physical and chemical phenomenons observable in real situations. A data assimilation method, the iterative ensemble Kalman smoother (IEnKS), has been applied to this model. It is an iterative 4D EnVar method which solves the full non-linear variational problem. This application validates 4D EnVar methods in the context of non-linear atmospheric chemistry, but also raises the first limits of such methods.After this experiment, results have been extended to a realistic atmospheric pollution prediction model. 4D EnVar methods, via the IEnKS, have once again shown their potential to take into account the non-linearity of the chemistry model in a controlled environment, with synthetic observations. However, the assimilation of real tropospheric ozone concentrations mitigates these results and shows how hard atmospheric chemistry data assimilation is. A strong model error is indeed attached to these models, stemming from multiple uncertainty sources. Two steps must be taken to tackle this issue.First of all, the data assimilation method used must be able to efficiently take into account the model error. However, most methods are developed under the assumption of a perfect model. To avoid this hypothesis, a new method has then been developed. Called IEnKF-Q, it expands the IEnKS to the model error framework. It has been validated on a low-order model, proving its superiority over data assimilation methods naively adapted to take into account model error.Nevertheless, such methods need to know the exact nature and amplitude of the model error which needs to be accounted for. Therefore, the second step is to use statistical tools to quantify this model error. The expectation-maximization algorithm, the naive and unbiased randomize-then-optimize algorithms, an importance sampling based on a Laplace proposal, and a Markov chain Monte Carlo simulation, potentially transdimensional, have been assessed, expanded, and compared to estimate the uncertainty on the retrieval of the source term of the Chernobyl and Fukushima-Daiichi nuclear power plant accidents.This thesis therefore improves the domain of 4D EnVar data assimilation by its methodological input and by paving the way to applying these methods on atmospheric chemistry models
Nkoumbou, Charles. „I. Étude géologique des Monts Roumpi : un ensemble plutonique et volcanique de la "Ligne du Cameroun"II. Données pétrologiques sur les néphélinites du Mont Etinde (Cameroun)“. Nancy 1, 1990. http://docnum.univ-lorraine.fr/public/SCD_T_1990_0460_NKOUMBOU.pdf.
Der volle Inhalt der QuelleLassalle, Pierre. „Etude du passage à l'échelle des algorithmes de segmentation et de classification en télédétection pour le traitement de volumes massifs de données“. Thesis, Toulouse 3, 2015. http://www.theses.fr/2015TOU30261/document.
Der volle Inhalt der QuelleRecent Earth observation spatial missions will provide very high spectral, spatial and temporal resolution optical images, which represents a huge amount of data. The objective of this research is to propose innovative algorithms to process efficiently such massive datasets on resource-constrained devices. Developing new efficient algorithms which ensure identical results to those obtained without the memory limitation represents a challenging task. The first part of this thesis focuses on the adaptation of segmentation algorithms when the input satellite image can not be stored in the main memory. A naive solution consists of dividing the input image into tiles and segment each tile independently. The final result is built by grouping the segmented tiles together. Applying this strategy turns out to be suboptimal since it modifies the resulting segments compared to those obtained from the segmentation without tiling. A deep study of region-merging segmentation algorithms allows us to develop a tile-based scalable solution to segment images of arbitrary size while ensuring identical results to those obtained without tiling. The feasibility of the solution is shown by segmenting different very high resolution Pléiades images requiring gigabytes to be stored in the memory. The second part of the thesis focuses on supervised learning methods when the training dataset can not be stored in the memory. In the frame of the thesis, we decide to study the Random Forest algorithm which consists of building an ensemble of decision trees. Several solutions have been proposed to adapt this algorithm for processing massive training datasets, but they remain either approximative because of the limitation of memory imposes a reduced visibility of the algorithm on a small portion of the training datasets or inefficient because they need a lot of read and write access on the hard disk. To solve those issues, we propose an exact solution ensuring the visibility of the algorithm on the whole training dataset while minimizing read and write access on the hard disk. The running time is analysed by varying the dimension of the training dataset and shows that our proposed solution is very competitive with other existing solutions and can be used to process hundreds of gigabytes of data
Daget, Nicolas. „Estimation d'ensemble des paramètres des covariances d'erreur d'ébauche dans un système d'assimilation variationnelle de données océaniques“. Toulouse 3, 2008. http://thesesups.ups-tlse.fr/251/1/Daget_Nicolas.pdf.
Der volle Inhalt der QuelleIn this thesis, an ensemble method has been designed and used to estimate parameters of the background-error covariance model in a three-dimensional variational data assimilation (3D-Var) system of a global version of the OPA ocean general circulation model. The ensemble is created by perturbing the surface forcing fields (windstress, fresh-water and heat flux) and the observations (temperature and salinity profiles) used in the assimilation process. This thesis work focused on the use of the ensemble to provide flow-dependent estimates of the background-error standard deviations (sigma b) for temperature and salinity. Cycled 3D-Var experiments were performed over the period 1993-2000 to test the sensitivity of the analyses to the ensemble sigma b formulation and to a simpler flow-dependent formulation of sigma b based on an empirical parameterization in terms of the vertical gradients of the background temperature and salinity fields. Both 3D-Var experiments produce a significant improvement in the fit to the temperature and salinity data compared to that from a control experiment in which no data were assimilated. Comparing innovation statistics from the two sigma b formulations shows that both formulations produce similar results except in the upper 150 m where the parameterized sigma b are slightly better. There, statistical consistency checks indicate that the ensemble sigma b are underestimated. The temperature and salinity error growth between cycles, however, is shown to be much reduced with the ensemble sigma b, suggesting that the analyses produced with the ensemble sigma b are in better balance than those produced with the parameterized sigma b. Sea surface height (SSH) anomalies in the northwest Atlantic and zonal velocities in the equatorial Pacific, which are fields not directly constrained by the observations, are clearly better with the ensemble sigma b than with the parameterized sigma b when compared to independent data. Comparisons with the control are less conclusive, and indicate that, while some aspects of those variables are improved (SSH anomalies and currents in the central and eastern Pacific), other aspects are degraded (SSH anomalies in the northwest Atlantic, currents in the western Pacific). .