Dissertations / Theses on the topic 'Technique d'apprentissage'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Technique d'apprentissage.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
Brunel, Stéphane Girard Philippe Zolghadri Marc. "Étude des activités collaboratives de conception en tant que situation d'apprentissage application à l'ingénierie des produits et à l'ingénierie didactique /." S. l. : S. n, 2008. http://ori-oai.u-bordeaux1.fr/pdf/2008/BRUNEL_STEPHANE_2008.pdf.
Full textMaranzana, Nicolas. "Amélioration de la performance en conception par l’apprentissage en réseau de la conception innovante." Strasbourg, 2009. http://www.theses.fr/2009STRA6223.
Full textIn the current economic context, price, quality and time are not any more the only competitive advantages since all the serious competitors are powerful on these three dimensions. In order to be different, companies have to innovate. To achieve this goal, the company has to be flexible and has to develop its intelligence. This leads to a better innovation process management and in particular a growth in its design activities effectiveness. This increase can be done thanks to two nonexclusive alternatives: on the one hand, a more extended control of various scientific knowledge, because of an increasing complexification of the technical systems; on the other hand, a rise in people competence on effective design methods, i. E. Leading to real modifications of existing systems, even to totally new systems. It is what we call innovative design. Various cases are possible for companies wishing to innovate. They can seek to innovate by themselves, to gather in corporate network to innovate together or to gather in network to learn how to innovate. This thesis problematic rests on the improvement of the performance in design by learning the innovative design in network. In order to limit resources, the partners gather their learning in order to learn together how to design better. The main contribution relates to the definition and the characterization of a network making it possible to join together companies wishing to develop their competences on new design methods
Delgoulet, Catherine. "La formation professionnelle des travailleurs vieillissants : composantes motivationnelles et modes d'apprentissage d'une technique de maintenance ferroviaire." Toulouse 2, 2000. http://www.theses.fr/2000TOU20008.
Full textSouto, Lopez Miguel. "Resserrer le dispositif européen de l'enseignement supérieur par les acquis d'apprentissage." Thesis, Lyon, École normale supérieure, 2015. http://www.theses.fr/2015ENSL0979.
Full textSince September 2014, the Belgian French-speaking higher education institutions must define learning outcomes for each of their programs. This thesis studies the processes which led to this obligation.The learning outcomes are understood as an educational concept with different uses. They mobilized by various European instruments: European qualification framework, Overarching framework of qualification in the European higher education area, quality assurance, ECTS and Diploma supplement labels. The objectives of these instruments are to enhance the transparency, the readability of the qualifications in order to support the mobility of the students and the workers.First, this research describes the expansion of the concept of competence in education. Secondly, it analyses the transposition of the European instruments in the French-speaking of Belgium. Thirdly, it presents a history of pedagogy in the three main universities.The data are discourses (educational texts, laws, institutional documents, international texts, interviews). Some of these discourses are analyzed with the TXM textometry software. Three theoretical frameworks are mobilized: actor-network theory and the concept of socio-technical networks (Callon & Latour), the Economies of worth and the idea of worlds of justification (Boltanski & Thevenot), and the concept European higher education apparatus (Croché, Charlier)
Troger, Vincent. "Histoire des centres d'apprentissage, 1939-1959 : les enjeux économiques, politiques et culturels de la constitution de l'enseignement technique court." Paris 4, 1991. http://www.theses.fr/1990PA040154.
Full textThe existence of apprenticeship centers, which are actually called "lycées professionnels", was linked both to long term phenomena and events connected with a series of exceptional situation. To succeed in doing the schooling of apprenticeship, the state was steadily supported by the employers of metallurgical industries who needed schools for the training of workers. But the history of apprenticeship centers was also influenced by the intense political tensions of the period between 1939 and 1948, which made workers-training a key factor and contributed towards building up the identity of this institution amid strong contradictions between their professional and socio-cultural aims
Gouzévitch. "Le transfert du savoir technique et scientifique et la construction de l'Etat russe (fin du XVe-début du XIXe siècle)." Paris 8, 2001. http://www.theses.fr/2001PA081925.
Full textBrunel, Stéphane. "Étude des activités collaboratives de conception en tant que situation d'apprentissage : application à l'ingénierie des produits et à l'ingénierie didactique." Phd thesis, Université Sciences et Technologies - Bordeaux I, 2008. http://tel.archives-ouvertes.fr/tel-00429635.
Full textMoulin, Sandrine. "Smac : un outil d'acquisition de connaissances pour des problèmes de conception." Toulouse, ENSAE, 1990. http://www.theses.fr/1990ESAE0001.
Full textChampciaux, Laurent. "Introduction de techniques d'apprentissage en modelisation declarative." Nantes, 1998. http://www.theses.fr/1998NANT2022.
Full textSenouci, Sidi-Mohammed. "Application de techniques d'apprentissage dans les réseaux mobiles." Paris 6, 2003. http://www.theses.fr/2003PA066485.
Full textBrezellec, Pierre. "Techniques d'apprentissage par explication et détections de similarités." Paris 13, 1992. http://www.theses.fr/1992PA132033.
Full textGirard, Didier. "Prevision de series temporelles par techniques locales d'apprentissage." Lyon, École normale supérieure (sciences), 1997. http://www.theses.fr/1997ENSL0074.
Full textAhriz, Roula Iness. "Application des techniques d'apprentissage à la géolocalisation par radio fingerprint." Phd thesis, Université Pierre et Marie Curie - Paris VI, 2010. http://pastel.archives-ouvertes.fr/pastel-00546952.
Full textAhriz, Iness. "Application des techniques d'apprentissage à la géolocalisation par radio fingerprint." Paris 6, 2010. http://www.theses.fr/2010PA066355.
Full textHo, Vinh Thanh. "Techniques avancées d'apprentissage automatique basées sur la programmation DC et DCA." Thesis, Université de Lorraine, 2017. http://www.theses.fr/2017LORR0289/document.
Full textIn this dissertation, we develop some advanced machine learning techniques in the framework of online learning and reinforcement learning (RL). The backbones of our approaches are DC (Difference of Convex functions) programming and DCA (DC Algorithm), and their online version that are best known as powerful nonsmooth, nonconvex optimization tools. This dissertation is composed of two parts: the first part studies some online machine learning techniques and the second part concerns RL in both batch and online modes. The first part includes two chapters corresponding to online classification (Chapter 2) and prediction with expert advice (Chapter 3). These two chapters mention a unified DC approximation approach to different online learning algorithms where the observed objective functions are 0-1 loss functions. We thoroughly study how to develop efficient online DCA algorithms in terms of theoretical and computational aspects. The second part consists of four chapters (Chapters 4, 5, 6, 7). After a brief introduction of RL and its related works in Chapter 4, Chapter 5 aims to provide effective RL techniques in batch mode based on DC programming and DCA. In particular, we first consider four different DC optimization formulations for which corresponding attractive DCA-based algorithms are developed, then carefully address the key issues of DCA, and finally, show the computational efficiency of these algorithms through various experiments. Continuing this study, in Chapter 6 we develop DCA-based RL techniques in online mode and propose their alternating versions. As an application, we tackle the stochastic shortest path (SSP) problem in Chapter 7. Especially, a particular class of SSP problems can be reformulated in two directions as a cardinality minimization formulation and an RL formulation. Firstly, the cardinality formulation involves the zero-norm in objective and the binary variables. We propose a DCA-based algorithm by exploiting a DC approximation approach for the zero-norm and an exact penalty technique for the binary variables. Secondly, we make use of the aforementioned DCA-based batch RL algorithm. All proposed algorithms are tested on some artificial road networks
Eychenne, Bertrand. "Le Colegio Militar de Bogota (1848-1884). La mise en place d'un enseignement supérieur scientifique et technique après l'indépendance de la Colombie." Thesis, Université Paris-Saclay (ComUE), 2018. http://www.theses.fr/2018SACLS176.
Full textAs Colombia becomes emancipated from the Spanish Crown in 1819, it has to strengthen its independence and is thus faced with a number of obstacles which come in the way of a stable scientific and technical education. By taking into account this postcolonial context and by focusing on the Colegio Militar of Bogota, a school of civil and military engineering between 1848 and 1884, this study describes the process which led to the foundation of such teachings and follows its evolution during the second half of the 19th century. The influence of power proves to be constant at the time and brings out such specificities as its mixed education curriculum, military and civil, its quest for utility or the absence of a selection process. Furthermore, the study of its curriculum reveals how its institutional vision influenced by party ideology shows through the teachings. It also demonstrates the ability of the various players to alter, at their level, this curriculum and manages to establish the continuity of its history despite the heterogeneity of their actions. Similarly, the analysis of the scientific and technical notions conveyed by this curriculum illustrates how, by adapting to the context, the diffusion of knowledge comes with scientific production. The decentralization entailed by our study leads us to consider the issue of transfer of knowledge, by showing how the study of an educational institution allows to tackle these mechanisms in their complexity. These circulations will be considered on different levels, locally with the regulating function of the school in the educational field, within the South American continent and toward the main hubs of scientific production in Europe and North America. Finally, this study sheds some light on the constitution and emergence of a professional group in Colombia namely, civil engineers, which is tightly linked to that of science teachers. By following the trail of former students of the Colegio Militar, we become acquainted with the strategies they adopt to ensure that both their profession and the right to practise it is acknowledged
Paré, Élaine. "Étude exploratoire de l'utilisation de la stratégie de schématisation chez les étudiantes et les étudiants de Techniques d'inhalothérapie." Sherbrooke : Université de Sherbrooke, 2000.
Find full textPorumbel, Daniel Cosmin. "Algorithmes Heuristiques et Techniques d'Apprentissage - Applications au Probleme de Coloration de Graphe." Phd thesis, Université d'Angers, 2009. http://tel.archives-ouvertes.fr/tel-00481253.
Full textPorumbel, Daniel Cosmin. "Algorithmes Heuristiques et Techniques d'Apprentissage : Applications au Problème de Coloration de Graphe." Phd thesis, Université d'Angers, 2009. http://tel.archives-ouvertes.fr/tel-00476541.
Full textPiworwarski, Benjamin. "Techniques d'apprentissage pour le traitement d'informations structurées : application à la recherche d'information." Paris 6, 2003. http://www.theses.fr/2003PA066567.
Full textVinter, Patricia. "Est-il possible et souhaitable d’enseigner la technique de la lecture indépendamment de sa finalité ? : elaboration d'une méthode de lecture qui différencie le décodage de la compréhension en phase d’apprentissage explicite et sa mise à l'épreuve en éducation prioritaire." Thesis, Lyon 2, 2015. http://www.theses.fr/2015LYO20002.
Full textSchool failure continues to increase in France, as well as the gap between the lowest and highest performing pupils (PISA 2003, 2006, 2009, 2012). Reading is a complex activity that requires mastering conjointly two skills, word identification and meaning understanding. Word identification is the main cause of reading difficulty: the code that binds oral and written information must be understood as being based on conventions, and the pupils who cannot access to symbols encounter difficulties in this understanding. In the present work, we have developed a new learning device that makes clear to children the relationships between oral and written syllables. This device comprises a representation of the writing system and material that enables to manipulate phonogrammes. To this end, we have included an additional step within a reading and writing method, the identification of pseudo-words, that is to say, of “signifieds” (plausible words) without “signifiers” (no corresponding referee or meaning). In order not to neglect the understanding dimension at the beginning of the learning phase, oral stories were presented to the children. They were extracted from an album of 30 chapters in which the heroine, a young witch, attributes meaning to these pseudo-words through her magic spells. In the training phase, the pseudo-words (associated with its signifier) are presented inside various reading - identification and understanding - and writing activities. Our specific training had the expected positive effects in unselected grade 1 elementary pupils from areas of prioritary education. These effects concerned mainly the identification and production of words, without adverse effects in other aspects of writing
Isaza, Narvaez Claudia Victoria. "Diagnostic par techniques d'apprentissage floues: concept d'une méthode de validation et d'optimisation des partitions." Phd thesis, INSA de Toulouse, 2007. http://tel.archives-ouvertes.fr/tel-00190884.
Full textIsaza, Narvaez Claudia Victoria. "Diagnostic par techniques d'apprentissage floues : conception d'une méthode de validation et d'optimisation des partitions." Toulouse, INSA, 2007. http://eprint.insa-toulouse.fr/archive/00000159/.
Full textThis work is in the field of the process diagnosis defined as the identification of process functional states. If obtaining a precise model of the process is delicate or impossible, the system knowledge can be extracted from the signals obtained during a normal or abnormal operation by including mechanisms of training. This knowledge is organized through a data space partition into clusters (representing the states of the system). Among the training techniques, those including fuzzy logic have the advantage of expressing the memberships of an individual to several classes, this makes possible to better know the real situation of the system and to envisage changes to failure states. Notwithstanding their adequate performances, their strong dependence on the initialization parameters is a difficulty for the training. This thesis proposes the improvement of these techniques, specifically our objective is the development of a method to validate and adapt automatically the partition of data space obtained by a fuzzy classification technique. This makes possible to find automatically an optimal partition in terms of clusters compactness and separation from only the membership matrix obtained by an initial classification. This method is thus an important help given to the process expert to establish the functional states in the implementation of a monitoring technique of a complex process. Its application is illustrated on academic examples and on the diagnosis of 3 chemical processes
Ganascia, Jean-Gabriel. "AGAPE et CHARADE deux techniques d'apprentissage symbolique appliquées à la construction de bases de connaissances /." Grenoble 2 : ANRT, 1987. http://catalogue.bnf.fr/ark:/12148/cb37605268r.
Full textGanascia, Jean-Gabriel. "AGAPE et CHARADE : deux techniques d'apprentissage symbolique appliquées à la construction de bases de connaissances." Paris 11, 1987. http://www.theses.fr/1987PA112135.
Full textHOCEINI, SAID. "Techniques d'Apprentissage par Renforcement pour le Routage Adaptatif dans les Réseaux de Télécommunication à Trafic Irrégulie." Phd thesis, Université Paris XII Val de Marne, 2004. http://tel.archives-ouvertes.fr/tel-00010430.
Full textNgo, Duy Hoa. "Amélioration de l'alignement d'ontologies par les techniques d'apprentissage automatique, d'appariement de graphes et de recherche d'information." Phd thesis, Université Montpellier II - Sciences et Techniques du Languedoc, 2012. http://tel.archives-ouvertes.fr/tel-00767318.
Full textHoceini, Said Amirat Yacine. "Techniques d'apprentissage par renforcement pour le routage adaptatif dans les réseaux de télécommunication à trafic régulier." Créteil : Université de Paris-Val-de-Marne, 2004. http://doxa.scd.univ-paris12.fr:80/theses/th0215018.pdf.
Full textHoceini, Said. "Techniques d'apprentissage par renforcement pour le routage adaptatif dans les réseaux de télécommunication à trafic régulier." Paris 12, 2004. https://athena.u-pec.fr/primo-explore/search?query=any,exact,990002150180204611&vid=upec.
Full textThe aim of this thesis is to propose an algorithmic approach, \vhich allo\vs to treat the problems of adaptive routing (AR) in telecommunication networks with irregular traffic. The analysis of the existing approaches has lead us to base our \vork on tlie Q-Routing (QR) algonthm. This algorithm uses a reinforcement learning technique vhich is based on Markov models. The efficiency of these routing approaches depends on information about the network load and the nature of data fiows. This information must be sufficient and relevant and l has to reflect the real network load during the decision making phase. To overcome drawbacks of techniques using QR, ve have proposed tivo AR algorithms. The first one, which is called Q-Neural Routing, is based on a stochastic neural model, used for parameter estimation and updating required for routing. In order to reduce the convergence time, a second approach is proposed: k-Shortest path Q-Routing. It is based on a multi-patlis routing technique combined with the QR algoritlim. In this case, the exploration space is limited to k-Best paths. The proposed algorithms are validated and compared to traditional approaches using the OPNET Simulator. Their efficiency, with respect to AR, is illustrated. In fact, these algorithms allow taking into account the network state in a hetter wav than the classical approaches do
Montoya-Obeso, Abraham. "Reconnaissance du patrimoine Mexicaine sous forme numérique par des réseaux d'apprentissage profond." Thesis, Bordeaux, 2020. http://www.theses.fr/2020BORD0064.
Full textIn Mexico, one of the priority technological problems is the preservation of cultural heritage in its digital form. In this research, the main interest is the ordering, management and identification of intangible cultural heritage in images. In computer vision, the integration of the Human Visual System (HVS) into automatic learning methods and classifiers has become an intensive research field for object recognition and content mining. The so-called saliency maps, are defined as a topographic representation of visual attention on a scene, modeling attention instantaneously and assigning a degree of interest to each pixel value on the image. Saliency maps proved to be very efficient to point out regions of interest in several tasks of visual content and its understanding. In this context, we focus on the integration of visual attention models in the training pipeline of Deep Neural Networks (DNNs) for the recognition of Mexican architectural structures. We consider the main contributions of this research are in the following areas of interest: • Specific purpose dataset: gathering data related to the topic is a key task to solve the problem of architectural classification. • Data selection: we use saliency prediction methods to select and crop context-relevant regions on images. • Visual attention modeling: we annotate images through a real task of image observation, we record eye-fixations with an eye-tracker system to build subjective saliency maps. • Visual attention integration: we integrate visual attention in deep neural networks in two ways; i) to filter out features in a saliency-based pooling layer and ii) in attention mechanisms. In this research, different essential components for the training of a neural network are tackled down with the aim of recognizing Mexican cultural content and extrapolating these findings to large-scale databases in similar classification tasks, such as in ImageNet. Finally, we show that the integration of visual attention models generated through a psycho-visual experiment allows to reduce training time and improve performances in terms of accuracy
Al, Hajj Mohamad Rami. "Reconnaissance hors ligne de mots manuscrits cursifs par l'utilisation de systèmes hybrides et de techniques d'apprentissage automatique." Paris, ENST, 2007. http://www.theses.fr/2007ENST0020.
Full textThe automatic offline recognition of handwritten words improves human-machine interaction. It is already used in many business office applications dealing with the automatic processing of documents such as automatic post sorting, and the verification and recognition of bank check amounts. The off line recognition of cursive handwritten words remains an open problem due to difficulties such as :handwriting normalization, word segmentation into compound components and the modeling of these components. The main objective of this thesis, is to propose, design, and implement a system for the automatic offline recognition of Arabic handwritten words. The proposed approach is analytical without explicit segmentation of words into compound characters, and it is based on the stochastic HMM approach (Hidden Markov models). The method is composed of two stages : a recognition stage based on different features, and a combination stage of three HMM-based classifiers. Each individual HMM classifier uses a sliding window with a specific inclination. Different combining strategies are tested, among them the Sum rule, the Majority Vote rule and the Borda Count rule. The best combination strategy consists of using a neural network-based combining classifier. The combination of these classifiers can better cope with the writing inclination, the erroneous positions of diacritical marks and points, and the overlapping of consecutive characters in handwritten words. The reference system based on the proposed method has shown best performance at the competition organized at ICDAR 2005, where a set of state-of art systems were compared and tested on the IFN/ENIT benchmark database
Saumtally, Anissa. "Economic catching-up, Technological progress and Intellectual property rights." Thesis, Bordeaux, 2017. http://www.theses.fr/2017BORD0829/document.
Full textThe objective of this thesis is to propose an answer to the question: Can intellectual property rights policies such as TRIPS be beneficial for developing countries and their catching-up process?To answer this question, we first look at the technological dynamics behind the catching-up process. The first chapter thus provides an empirical and analytical update on the catching-up and falling behind model by Verspagen (1991), which focuses on studying the role of the innovation and imitation dynamics in the catching up process. Mainly, we find that while the innovation dynamic is important for the catching-up process, the imitation dynamic is necessary to ensure that countries build solid capabilities that will enable them to prosper. The efficiency of the imitation dynamics is dependent on policy factors that make up the learning capability of firms and ensure firms succeed assimilating knowledge.The second chapter focuses on understanding the way those technological transfers from developed to developing countries can occur, we focus on studying the mechanisms behind two main channels, that is international trade and FDIs, which represent the main form of North-South interactions studied in the literature. From this chapter we conclude that there is a rich diversity of complex mechanisms.In the third chapter, we thus build an agent-based model (ABM) to represent those North-South interactions and their complexities, with an evolutionary economics approach. The model allows us to study a particular mechanism: transfers through the local labour mobility, a channel seldom discussed in the literature. This allows us to study the impact FDI may have on development and catching-up outcomes. We find that while FDI from developed countries can, under the right conditions, encourage technological transfers and thus catching up, there are potential negative effects on local industries, in particular in countries largely behind.The final chapter proposes an extension of the model that introduces patents, in order to answer the main question. We find that while patents help motivate northern firms to disclose their technology and thus facilitate development, those firms would require a perfect level of enforcement that will be too harsh on local firms, block imitations and also severely hinder the southern firms’ innovative efforts, while generating limited gains for northern firms
Jacquemart, Nathalie. "Transmission et techniques d'apprentissage d'un savoir traditionnel : étude ethnolinguistique et ethnomusicologique de la musique de Gamelan (Java central)." Paris 5, 1997. http://www.theses.fr/1997PA05H081.
Full textGamelans are traditional javanese music orchestras, essentially composed of metallophones and gongs. They play pure music, accompaniement for dances, shadows and actors theatre. This music plays an essential function in cultural life of central Java: weddings, festivities, competitions, relaxation. The numerous orchestras assemble people from every parts of population. Gamelan takes place in public and private entreprises, district and village associations. Radio, hundred of cassettes, concerts prove population's fancy. Linked to this development, transmission has stronglu evolved. Ciffer notations exist since the beginning of this century. They are used by everyone. The first conservatory of Indonesia was created in Surakarta, by 1950. Lessons are also given, for the numerous amateur groups, by musicians formed or not at school. But we can observe that their pedagogic methodes are very similar. If music tends to be fixed, teachers use, nevertheless, learning methods which permit to preserve richness and diversity of performance. This study brings knowledge on transmission of tradional music (rarely studied), javanese society and culture (music reveals general characteristics of culture), and on specificity of gamelan's musical practices through their present evolution. It has implied crossed investment of linguistics (ethnolinguistic methodology), anthropology (general conception of theme of learning) and ethnomusicology (bring out a system by preferential way of pedagogy, and organology). We can bring out that learning, as subject of study, permits to reach systems, in their structural, historical and sociocultural dimensions
Girard, Nicolas. "Approches d'apprentissage et géométrique pour l'extraction automatique d'objets à partir d'images de télédétection." Thesis, Université Côte d'Azur, 2020. https://tel.archives-ouvertes.fr/tel-03177997.
Full textCreating a digital double of the Earth in the form of maps has many applications in e.g. autonomous driving, automated drone delivery, urban planning, telecommunications, and disaster management. Geographic Information Systems (GIS) are the frameworks used to integrate geolocalized data and represent maps. They represent shapes of objects in a vector representation so that it is as sparse as possible while representing shapes accurately, as well as making it easier to edit than raster data. With the increasing amount of satellite and aerial images being captured every day, automatic methods are being developed to transfer the information found in those remote sensing images into Geographic Information Systems. Deep learning methods for image segmentation are able to delineate the shapes of objects found in images however they do so with a raster representation, in the form of a mask. Post-processing vectorization methods then convert that raster representation into a vector representation compatible with GIS. Another challenge in remote sensing is to deal with a certain type of noise in the data, which is the misalignment between different layers of geolocalized information (e.g. between images and building cadaster data). This type of noise is frequent due to various errors introduced during the processing of remote sensing data. This thesis develops combined learning and geometric approaches with the purpose to improve automatic GIS mapping from remote sensing images.We first propose a method for correcting misaligned maps over images, with the first motivation for them to match, but also with the motivation to create remote sensing datasets for image segmentation with alignment-corrected ground truth. Indeed training a model on misaligned ground truth would not lead to great performance, whereas aligned ground truth annotations will result in better models. During this work we also observed a denoising effect of our alignment model and use it to denoise a misaligned dataset in a self-supervised manner, meaning only the misaligned dataset was used for training.We then propose a simple approach to use a neural network to directly output shape information in the vector representation, in order to by-pass the post-processing vectorization step. Experimental results on a dataset of solar panels show that the proposed network succeeds in learning to regress polygon coordinates, yielding directly vectorial map outputs. Our simple method is limited to predicting polygons with a fixed number of vertices though.While more recent methods for learning directly in the vector representation do not have this limitation, they still have other limitations in terms of the type of object shapes they can predict. More complex topological cases such as objects with holes or buildings touching each other (with a common wall which is very typical of European city centers) are not handled by these fully deep learning methods. We thus propose a hybrid approach alleviating those limitations by training a neural network to output a segmentation probability map as usual and also to output a frame field aligned with the contours of detected objects (buildings in our case). That frame field constitutes additional shape information learned by the network. We then propose our highly parallelizable polygonization method for leveraging that frame field information to vectorize the segmentation probability map efficiently. Because our polygonization method has access to additional information in the form of a frame field, it can be less complex than other advanced vectorization methods and is thus faster. Lastly, requiring an image segmentation network to also output a frame field only adds two convolutional layers and virtually does not increase inference time, making the use of a frame field only beneficial
Diallo, Boubacar. "Mesure de l'intégrité d'une image : des modèles physiques aux modèles d'apprentissage profond." Thesis, Poitiers, 2020. http://www.theses.fr/2020POIT2293.
Full textDigital images have become a powerful and effective visual communication tool for delivering messages, diffusing ideas, and proving facts. The smartphone emergence with a wide variety of brands and models facilitates the creation of new visual content and its dissemination in social networks and image sharing platforms. Related to this phenomenon and helped by the availability and ease of use of image manipulation softwares, many issues have arisen ranging from the distribution of illegal content to copyright infringement. The reliability of digital images is questioned for common or expert users such as court or police investigators. A well known phenomenon and widespread examples are the "fake news" which oftenly include malicious use of digital images.Many researchers in the field of image forensic have taken up the scientific challenges associated with image manipulation. Many methods with interesting performances have been developed based on automatic image processing and more recently the adoption of deep learning. Despite the variety of techniques offered, performance are bound to specific conditions and remains vulnerable to relatively simple malicious attacks. Indeed, the images collected on the Internet impose many constraints on algorithms questioning many existing integrity verification techniques. There are two main peculiarities to be taken into account for the detection of a falsification: one is the lack of information on pristine image acquisition, the other is the high probability of automatic transformations linked to the image-sharing platforms such as lossy compression or resizing.In this thesis, we focus on several of these image forensic challenges including camera model identification and image tampering detection. After reviewing the state of the art in the field, we propose a first data-driven method for identifying camera models. We use deep learning techniques based on convolutional neural networks (CNNs) and develop a learning strategy considering the quality of the input data versus the applied transformation. A family of CNN networks has been designed to learn the characteristics of the camera model directly from a collection of images undergoing the same transformations as those commonly used on the Internet. Our interest focused on lossy compression for our experiments, because it is the most used type of post-processing on the Internet. The proposed approach, therefore, provides a robust solution to compression for camera model identification. The performance achieved by our camera model detection approach is also used and adapted for image tampering detection and localization. The performances obtained underline the robustness of our proposals for camera model identification and image forgery detection
Jobin, Isabelle. "La pédagogie Freinet à l'école optionnelle Yves-Prévost : un parcours d'apprentissage professionnel en communauté de pratique." Thesis, Université Laval, 2013. http://www.theses.ulaval.ca/2013/29660/29660.pdf.
Full textBrahmi, Djamel Fertil Bernard. "Quantification de la progression virale dans les rétinopathies à CMV par des techniques d'analyse d'images fondées sur des méthodes d'apprentissage par l'exemple." [S. l.] : [S. n.], 2001. http://www.imed.jussieu.fr/niveau1/vie/actu_U494/TheseDB.html.
Full textBrahmi, Djamel. "Quantification de la progression virale dans les rétinopathies à CMV par des techniques d'analyse d'images fondées sur des méthodes d'apprentissage par l'exemple." Paris 5, 2001. http://www.theses.fr/2001PA05CD04.
Full textChampagne, André. "Analyse évaluative des objectifs d'apprentissage de sessions de formation pré-départ pour conseillers techniques oeuvrant en coopération internationale : le cas de l'ACDI." Thesis, University of Ottawa (Canada), 1994. http://hdl.handle.net/10393/6624.
Full textArgaud, Henri-Claude. "Problèmes et milieux a-didactiques, pour un processus d'apprentissage en géométrie plane à l'école élémentaire, dans les environnements papier-crayon et Cabri-géomètre." Université Joseph Fourier (Grenoble), 1998. http://www.theses.fr/1998GRE10129.
Full textDennouni, Nassim. "Orchestration des activités d’apprentissage mobile." Thesis, Lille 1, 2016. http://www.theses.fr/2016LIL10001/document.
Full textMobile learning has become a topic of interest because it involves many areas of research concerning usage contexts and complex technology. Indeed, mobile learning is has been recognized for its ability to motivate learners because they can construct their own knowledge by collaborating with others. In this context, the orchestration of mobile learning allows real-time management and contextualization of activities to do but this results in significant costs of organization. In addition, centralized orchestration is not adapted to the context of our mobile scenario because the learner must be able to keep some control over their choices of learning.In this thesis, we present a new style of recommendation for a dynamic orchestration of learning activities based on the location of the learners and the history of the visit. This technique is based on a collaborative filtering that exploits prior activity of the learners and that respects the educational and location constraints. Our approach is based on the mode of operation of the Swarm Intelligence (ACO algorithm) for the implementation of our system of recommendation. Besides the simulations that are used to compare the different variants of recommendations, the validation of the SAMSSP system goes through the experimentation of the two prototypes of campus visit
Bonfils, Philippe. "Dispositifs socio-techniques et mondes persistants : quelles médiations pour quelle communication dans un contexte situé ?" Phd thesis, Université du Sud Toulon Var, 2007. http://tel.archives-ouvertes.fr/tel-00257247.
Full textPetrov, Aleksandar. "Understanding the relationships between aesthetic properties of shapes and geometric quantities of free-form curves and surfaces using Machine Learning Techniques." Thesis, Paris, ENSAM, 2016. http://www.theses.fr/2016ENAM0007/document.
Full textToday on the market we can find a large variety of different products and differentshapes of the same product and this great choice overwhelms the customers. It is evident that the aesthetic appearance of the product shape and its emotional affection will lead the customers to the decision for buying the product. Therefore, it is very important to understand the aesthetic proper-ties and to adopt them in the early product design phases. The objective of this thesis is to propose a generic framework for mapping aesthetic properties to 3D freeform shapes, so as to be able to extract aesthetic classification rules and associated geometric properties. The key element of the proposed framework is the application of the Data Mining (DM) methodology and Machine Learning Techniques (MLTs) in the mapping of aesthetic properties to the shapes. The application of the framework is to investigate whether there is a common judgment for the flatness perceived from non-professional designers. The aim of the framework is not only to establish a structure for mapping aesthetic properties to free-form shapes, but also to be used as a guided path for identifying a mapping between different semantics and free-form shapes. The long-term objective of this work is to define a methodology to efficiently integrate the concept of Affective Engineering in the Industrial Designing
Liakopoulos, Nikolaos. "Machine Learning Techniques for Online Resource Allocation in Wireless Networks." Electronic Thesis or Diss., Sorbonne université, 2019. http://www.theses.fr/2019SORUS529.
Full textTraditionally, network optimization is used to provide good configurations in real network system problems based on mathematical models and statistical assumptions. Recently, this paradigm is evolving, fueled by an explosion of availability of data. The modern trend in networking problems is to tap into the power of data to extract models and deal with uncertainty. This thesis proposes algorithmic frameworks for wireless networks, based both on classical or data-driven optimization and machine learning. We target two use cases, user association and cloud resource reservation.The baseline approach for user association, connecting wireless devices to the base station that provides the strongest signal, leads to very inefficient configurations even in current wireless networks. We focus on tailoring user association based on resource efficiency and service requirement satisfaction, depending on the underlying network demand. We first study distributed user association with priority QoS guarantees, then scalable centralized load balancing based on computational optimal transport and finally robust user association based on approximate traffic prediction.Moving to the topic of cloud resource reservation, we develop a novel framework for resource reservation in worst-case scenaria, where the demand is engineered by an adversary aiming to harm our performance. We provide policies that have ``no regret'' and guarantee asymptotic feasibility in budget constraints under such workloads. More importantly we expand to a general framework for online convex optimization (OCO) problems with long term budget constraints complementing the results of recent literature in OCO
Takouachet, Nawel. "Utilisation de critères perceptifs pour la déterminatin d'une condition d'arrêt dans les méthodes d'illumination globale." Littoral, 2009. http://www.theses.fr/2009DUNK0229.
Full textThe thesis focused on modelsnof realistic images rendering, especially unbiased algorithms of global illumination. Their interest is to calculate precisely illumination solution which allows to produce realistic images. However, these methods are prone to visual noise du to the stochastic nature of the underlying methods. This noise can be reduced by increasing the number of computed samples but simultaneously increasing the computation times. In this thesis we have been interested in searching for automatic stopping criteria for these algorithms. More specifically we focused of perceptual criteria allowing visible noise to be detected through any image. After an overview of the different methods used to render images, the problem of the integration of the perceptual models is considered. We use knowledge of the human visual system to guide image rendering algorithms. In a second step, two methodologies are proposed. They are based respectively on a human visual model and a supervised learning approach. We calibrate these two methods through experimental data obtained from human observers. By comparing our two methods we show that one based on supervised learning has more advantages : it requires less additional memory and computation can be distributed heterogeneously across the image, focusing on noisy areas
Couland, Quentin. "Contribution à l'apprentissage humain de gestes à l'aide de techniques de clustering pour l’analyse de mouvements capturés." Thesis, Le Mans, 2020. http://www.theses.fr/2020LEMA1011.
Full textThis PhD thesis lies at the crossroads of the Technology Enhanced Learning (TEL) and human motion learning fields. A lot of TEL systems for motion learning already exist, and they are used in numerous application domains. While efficient in the task they were designed for, they are usually ad-hoc by design, focusing on a specific task and learning context. Reusing such systems in other learning contexts is impossible or requires a heavy re-engineering process. The design of TEL systems for motion learning, expandable beyond the task they were created for and needing a minimal amount of re-engineering, represents a challenge from which arise several technical issues and scientific questions. To tackle these challenges, the Motion Learning Analytics (MLA) system was developed. This system was tested on throwing motions through three different experimentations, designed to test four aspects of the system : (i) the possibility to achieve a good separation of the motions into multiple groups orresponding to different throwing strategies, (ii), the possibility to achieve a good separation corresponding to the degree of success of these motions, (iii), the integration of the expert's observations needs as criteria for the separation and evaluation of the learner's progression and (iv), an analysis on the ability of the system to provide an efficient and relevant assistance to the expert in order to improve the learner's gesture
Lachand-Pascal, Valentin. "Approche centrée activité pour la conception et l'orchestration d'activités numériques en classe." Thesis, Lyon, 2020. http://www.theses.fr/2020LYSEI089.
Full textThe quantity and variety of digital devices available in schools is constantly increasing. However, educational uses have not followed this evolution. The limited use of digital may be explained by the difficulty in creating digital activities. We propose to combine concepts from research related to the creation and conduction of educational activities in the classroom, as well as work related to the conduction of digital activities in a more general way. We present the contributions and limitations of these two conceptual frameworks for the management of digital educational activities. We conducted interviews with teachers to understand how they create their digital activities and how they conduct them. When conducting activities in the classroom, teachers mainly encounter technical problems (unstable network, hardware limitations) and have to adapt their plans to deal with these problems. We propose an adaptable architecture to solve the technical problems. Our architecture allows the conduction of digital activities by taking into account the different constraints: the activities work with or without connection, on the different devices present in the classroom, and the architecture adapts to the infrastructures of the schools. This architecture is materialized in Toccata, an orchestration system allowing the creation and implementation of digital activities in the classroom. Toccata implements the design recommendations we identified. Finally, we identified interaction strategies to help teachers in the management of digital devices in the classroom. Through observations of middle school classrooms, we identified two main classes of tasks: content sharing and remote control of devices. Using an elicitation study, we found that control tasks are easier to perform than content sharing tasks, especially using a device worn like a connected watch. We found that the selection of content sharing devices remained particularly complex in terms of interaction. Our results open new possibilities for managing digital devices in the classroom. However, work is still needed on sharing, reusing, and redesigning digital activities
Grim-Yefsah, Malika. "Gestion des connaissances et externalisation informatique. Apports managériaux et techniques pour l'amélioration du processus de transition : Cas de l’externalisation informatique dans un EPST." Thesis, Paris 9, 2012. http://www.theses.fr/2012PA090047/document.
Full textThe research of this thesis deals with the issue of knowledge transfer during the transition process of an IT project outsourced in EPST. In particular, How to transfer knowledge, experience and routines related to outsourced activities from outgoing team to a new incoming team? We focus on the transition due to its significance for outsourcing success, its complexity and theoretical richness, and its limited current understanding. We chose to approach this problem through knowledge management. In the first part of this thesis, based on the Goal-Question-Metric paradigm, we propose an approach for the definition of quality metrics covering the given operational requirements. The metrics we define take tacit knowledge into account, using information from the structural analysis of an informal network. In a second phase of this research, we developed a method, relying on capitalization on knowledge and theoretical mechanisms of knowledge transfer, and a tool to implement this process of knowledge transfer
Pawlowski, Filip igor. "High-performance dense tensor and sparse matrix kernels for machine learning." Thesis, Lyon, 2020. http://www.theses.fr/2020LYSEN081.
Full textIn this thesis, we develop high performance algorithms for certain computations involving dense tensors and sparse matrices. We address kernel operations that are useful for machine learning tasks, such as inference with deep neural networks (DNNs). We develop data structures and techniques to reduce memory use, to improve data locality and hence to improve cache reuse of the kernel operations. We design both sequential and shared-memory parallel algorithms. In the first part of the thesis we focus on dense tensors kernels. Tensor kernels include the tensor--vector multiplication (TVM), tensor--matrix multiplication (TMM), and tensor--tensor multiplication (TTM). Among these, TVM is the most bandwidth-bound and constitutes a building block for many algorithms. We focus on this operation and develop a data structure and sequential and parallel algorithms for it. We propose a novel data structure which stores the tensor as blocks, which are ordered using the space-filling curve known as the Morton curve (or Z-curve). The key idea consists of dividing the tensor into blocks small enough to fit cache, and storing them according to the Morton order, while keeping a simple, multi-dimensional order on the individual elements within them. Thus, high performance BLAS routines can be used as microkernels for each block. We evaluate our techniques on a set of experiments. The results not only demonstrate superior performance of the proposed approach over the state-of-the-art variants by up to 18%, but also show that the proposed approach induces 71% less sample standard deviation for the TVM across the d possible modes. Finally, we show that our data structure naturally expands to other tensor kernels by demonstrating that it yields up to 38% higher performance for the higher-order power method. Finally, we investigate shared-memory parallel TVM algorithms which use the proposed data structure. Several alternative parallel algorithms were characterized theoretically and implemented using OpenMP to compare them experimentally. Our results on up to 8 socket systems show near peak performance for the proposed algorithm for 2, 3, 4, and 5-dimensional tensors. In the second part of the thesis, we explore the sparse computations in neural networks focusing on the high-performance sparse deep inference problem. The sparse DNN inference is the task of using sparse DNN networks to classify a batch of data elements forming, in our case, a sparse feature matrix. The performance of sparse inference hinges on efficient parallelization of the sparse matrix--sparse matrix multiplication (SpGEMM) repeated for each layer in the inference function. We first characterize efficient sequential SpGEMM algorithms for our use case. We then introduce the model-parallel inference, which uses a two-dimensional partitioning of the weight matrices obtained using the hypergraph partitioning software. The model-parallel variant uses barriers to synchronize at layers. Finally, we introduce tiling model-parallel and tiling hybrid algorithms, which increase cache reuse between the layers, and use a weak synchronization module to hide load imbalance and synchronization costs. We evaluate our techniques on the large network data from the IEEE HPEC 2019 Graph Challenge on shared-memory systems and report up to 2x times speed-up versus the baseline
Baskaya, Elgiz. "Détection & diagnostic de pannes pour les drones utilisant la machine learning." Thesis, Toulouse 3, 2019. http://www.theses.fr/2019TOU30043.
Full textThis new era of small UAVs currently populating the airspace introduces many safety concerns, due to the absence of a pilot onboard and the less accurate nature of the sensors. This necessitates intelligent approaches to address the emergency situations that will inevitably arise for all classes of UAV operations as defined by EASA (European Aviation Safety Agency). Hardware limitations for these small vehicles point to the utilization of analytical redundancy, rather than to the usual practice of hardware redundancy in manned aviation. In the course of this study, machine learning practices are implemented in order to diagnose faults on a small fixed-wing UAV to avoid the burden of accurate modeling needed in model-based fault diagnosis. A supervised classification method, SVM (Support Vector Machines) is used to classify the faults. The data used to diagnose the faults are gyro and accelerometer measurements. The idea to restrict the data set to accelerometer and gyro measurements is to check the method's classification ability, with a small and inexpensive chip set and without the need to access the data from the autopilot, such as the control input information. This work addresses the faults in the control surfaces of a UAV. More specifically, the faults considered are the control surface stuck at an angle and the loss of effectiveness.First, a model of an aircraft is simulated. This model is not used for the design of Fault Detection and Diagnosis (FDD) algorithms, but is instead utilized to generate data. Simulated data are used instead of flight data in order to isolate the probable effects of the controller on the diagnosis, which may complicate a preliminary study on FDD for drones. The results show that for simulated measurements, SVM gives very accurate results on the classification of the loss of effectiveness faults on the control surfaces. These promising results call for further investigation so as to assess SVM performance on fault classification with flight data. Real flights were arranged to generate faulty flight data by manipulating the open source autopilot, Paparazzi. All data and the code are available in the code sharing and versioning system, Github. Training is held offline due to the need for labeled data and the computational burden of the tuning phase of the classifiers. Results show that from the flight data, SVM yields an F1 score of 0.98 for the classification of control surface stuck faults. For the loss of efficiency faults, some feature engineering, involving the addition of past measurements is needed in order to attain the same classification performance. A promising result is discovered when spinors are used as features instead of angular velocities. Results show that by using spinors for classification, there is a vast improvement in classification accuracy, especially when the classifiers are untuned. Using spinors and a Gaussian Kernel, an untuned classifier gives an F1 score of 0.9555, which was 0.2712 when gyro measurements were used as features. In summary, this work shows that SVM gives a satisfactory performance for the classification of faults on the control surfaces of a drone using flight data