Tesis sobre el tema "Apprentissge automatique"
Crea una cita precisa en los estilos APA, MLA, Chicago, Harvard y otros
Consulte los 50 mejores tesis para su investigación sobre el tema "Apprentissge automatique".
Junto a cada fuente en la lista de referencias hay un botón "Agregar a la bibliografía". Pulsa este botón, y generaremos automáticamente la referencia bibliográfica para la obra elegida en el estilo de cita que necesites: APA, MLA, Harvard, Vancouver, Chicago, etc.
También puede descargar el texto completo de la publicación académica en formato pdf y leer en línea su resumen siempre que esté disponible en los metadatos.
Explore tesis sobre una amplia variedad de disciplinas y organice su bibliografía correctamente.
Chouchene, Sarah. "Applications de l’intelligence artificielle à l'étude de la turbulence plasma en fusion nucléaire et aux plasmas d’arc en régime DC". Electronic Thesis or Diss., Université de Lorraine, 2024. http://www.theses.fr/2024LORR0265.
Texto completoThe aim of the work presented in this thesis is to explore the potential contribution of artificial intelligence (AI) image analysis methods to the analysis of videos from various plasma research domains. Two areas are studied: turbulence in nuclear fusion plasmas, and the dynamics of dipolar arcs in the high-voltage DC regime.This thesis mainly explores the application of AI to detect and track macroscopic turbulent structures visible in the form of plasma filaments at the edge of fusion reactors. To this end, AI models are applied to imaging data from the COMPASS tokamak captured by an ultra-fast camera filming up to 1 million frames per second in order to resolve the dynamics of the filaments. Different supervised learning models are employed, after training on labelled datasets to improve detection accuracy, to study the motion of plasma filaments and their interactions with each other. The methods presented in this thesis achieve a detection accuracy of 99 %, and automatically recognise different types of mutual interactions leading to coalescence, splitting or changes in filament trajectories. Comparisons with conventional analysis methods such as detection by segmentation and Kalman filter tracking show that the AI methods tested offer substantial gains in terms of accuracy and analysis speed, while reducing user bias.In the field of dipolar arcs, an unsupervised learning technique is used to detect bubbles of liquid metal forming on the surface of electrodes under high-voltage direct current (HVDC) conditions. The unsupervised models, which are lighter and operate without labelled data, are effective in identifying events that could be similar to anomalies. The results obtained contribute to a better understanding of arc noise, and pave the way for real-time data analysis for the implementation of protection systems.This research demonstrates the relevance of further developing AI methods to advance our understanding of the complex dynamics of nuclear fusion plasmas and arc plasmas. By reducing or eliminating analysis biases linked to human intervention, these methods can also help to improve comparisons between data from different experiments or simulations. The gains in analysis speed are not always very significant, but there are margins for optimisation that open up interesting prospects for improving the control of plasmas, whether cold or nuclear fusion
Tommasi, Marc. "Structures arborescentes et apprentissage automatique". Habilitation à diriger des recherches, Université Charles de Gaulle - Lille III, 2006. http://tel.archives-ouvertes.fr/tel-00117063.
Texto completoÀ la base de ce travail se trouve la question de l'accès et de la manipulation automatique d'informations au format XML au sein d'un réseau d'applications réparties dans internet. La réalisation de ces applications est toujours du ressort de programmeurs spécialistes d'XML et reste hors de portée de l'utilisateur final. De plus, les développements récents d'internet poursuivent l'objectif d'automatiser les communications entre applications s'échangeant des flux de données XML. Le recours à des techniques d'apprentissage automatique est une réponse possible à cette situation.
Nous considèrons que les informations sont décrites dans un langage XML, et dans la perspective de ce mémoire, embarquées dans des données structurées sous forme arborescente. Les applications sont basées alors sur des opérations élémentaires que sont l'interrogation ou les requêtes dans ces documents arborescents ou encore la transformation de tels documents.
Nous abordons alors la question sous l'angle de la réalisation automatique de programmes d'annotation d'arbres, permettant de dériver des procédures de transformation ou d'exécution de requêtes. Le mémoire décrit les contributions apportées pour la manipulation et l'apprentissage d'ensembles d'arbres d'arité non bornée (comme le sont les arbres XML), et l'annotation par des méthodes de classification supervisée ou d'inférence statistique.
Pintado, Michel. "Apprentissage et demonstration automatique des theoremes". Paris 6, 1994. http://www.theses.fr/1994PA066670.
Texto completoNatowicz, René. "Apprentissage symbolique automatique en reconnaissance d'images". Paris 11, 1987. http://www.theses.fr/1987PA112301.
Texto completoNatowicz, René. "Apprentissage symbolique automatique en reconnaissance d'images". Grenoble 2 : ANRT, 1987. http://catalogue.bnf.fr/ark:/12148/cb37608385b.
Texto completoCandillier, Laurent Gilleron Rémi. "Apprentissage automatique de profils de lecteurs". [S.l.] : [s.n.], 2001. http://www.univ-lille1.fr/bustl-grisemine/pdf/memoires/A2001-6.pdf.
Texto completoBayoudh, Sabri. "Apprentissage par proportion analogique". Rennes 1, 2007. ftp://ftp.irisa.fr/techreports/theses/2007/bayoudh.pdf.
Texto completoThe work presented in this thesis lies within the scope of reasoning by analogy. We are interested in the analogical proportion (A is to B as C is to D) and we describe its use and especially its contribution in machine learning. Firstly, we are interested in defining exact analogical proportions. Then, we tackle the problem of defining a new concept, the analogical dissimilarity which is a measure of how close four objects are from being in analogical proportion, including the case where the objects are sequences. After having defined the analogical proportion, the analogical dissimilarity and the approximate resolution of analogical equations, we describe two algorithms that make these concepts operational for numerical or symbolic objects and sequences of these objects. We show their use through two practical cases : the first is a problem of learning a classification rule on benchmarks of binary and nominal data ; the second shows how the generation of new sequences by solving analogical equations enables a handwritten character recognition system to rapidly be adapted to a new writer
Suchier, Henri-Maxime. "Nouvelles contributions du boosting en apprentissage automatique". Phd thesis, Université Jean Monnet - Saint-Etienne, 2006. http://tel.archives-ouvertes.fr/tel-00379539.
Texto completoLe boosting, et son algorithme AdaBoost, est une méthode ensembliste très étudiée depuis plusieurs années : ses performances expérimentales remarquables reposent sur des fondements théoriques rigoureux. Il construit de manière adaptative et itérative des hypothèses de base en focalisant l'apprentissage, à chaque nouvelle itération, sur les exemples qui ont été difficiles à apprendre lors des itérations précédentes. Cependant, AdaBoost est relativement inadapté aux données du monde réel. Dans cette thèse, nous nous concentrons en particulier sur les données bruitées, et sur les données hétérogènes.
Dans le cas des données bruitées, non seulement la méthode peut devenir très lente, mais surtout, AdaBoost apprend par coeur les données, et le pouvoir prédictif des hypothèses globales générées, s'en trouve extrêmement dégradé. Nous nous sommes donc intéressés à une adaptation du boosting pour traiter les données bruitées. Notre solution exploite l'information provenant d'un oracle de confiance permettant d'annihiler les effets dramatiques du bruit. Nous montrons que notre nouvel algorithme conserve les propriétés théoriques du boosting standard. Nous mettons en pratique cette nouvelle méthode, d'une part sur des données numériques, et d'autre part, de manière plus originale, sur des données textuelles.
Dans le cas des données hétérogènes, aucune adaptation du boosting n'a été proposée jusqu'à présent. Pourtant, ces données, caractérisées par des attributs multiples mais de natures différentes (comme des images, du son, du texte, etc), sont extrêmement fréquentes sur le web, par exemple. Nous avons donc développé un nouvel algorithme de boosting permettant de les utiliser. Plutôt que de combiner des hypothèses boostées indépendamment, nous construisons un nouveau schéma de boosting permettant de faire collaborer durant l'apprentissage des algorithmes spécialisés sur chaque type d'attribut. Nous prouvons que les décroissances exponentielles des erreurs sont toujours assurées par ce nouveau modèle, aussi bien d'un point de vue théorique qu'expérimental.
Paumard, Marie-Morgane. "Résolution automatique de puzzles par apprentissage profond". Thesis, CY Cergy Paris Université, 2020. http://www.theses.fr/2020CYUN1067.
Texto completoThe objective of this thesis is to develop semantic methods of reassembly in the complicated framework of heritage collections, where some blocks are eroded or missing.The reassembly of archaeological remains is an important task for heritage sciences: it allows to improve the understanding and conservation of ancient vestiges and artifacts. However, some sets of fragments cannot be reassembled with techniques using contour information or visual continuities. It is then necessary to extract semantic information from the fragments and to interpret them. These tasks can be performed automatically thanks to deep learning techniques coupled with a solver, i.e., a constrained decision making algorithm.This thesis proposes two semantic reassembly methods for 2D fragments with erosion and a new dataset and evaluation metrics.The first method, Deepzzle, proposes a neural network followed by a solver. The neural network is composed of two Siamese convolutional networks trained to predict the relative position of two fragments: it is a 9-class classification. The solver uses Dijkstra's algorithm to maximize the joint probability. Deepzzle can address the case of missing and supernumerary fragments, is capable of processing about 15 fragments per puzzle, and has a performance that is 25% better than the state of the art.The second method, Alphazzle, is based on AlphaZero and single-player Monte Carlo Tree Search (MCTS). It is an iterative method that uses deep reinforcement learning: at each step, a fragment is placed on the current reassembly. Two neural networks guide MCTS: an action predictor, which uses the fragment and the current reassembly to propose a strategy, and an evaluator, which is trained to predict the quality of the future result from the current reassembly. Alphazzle takes into account the relationships between all fragments and adapts to puzzles larger than those solved by Deepzzle. Moreover, Alphazzle is compatible with constraints imposed by a heritage framework: at the end of reassembly, MCTS does not access the reward, unlike AlphaZero. Indeed, the reward, which indicates if a puzzle is well solved or not, can only be estimated by the algorithm, because only a conservator can be sure of the quality of a reassembly
Jalam, Radwan. "Apprentissage automatique et catégorisation de textes multilingues". Lyon 2, 2003. http://theses.univ-lyon2.fr/documents/lyon2/2003/jalam_r.
Texto completoSani, Amir. "Apprentissage automatique pour la prise de décisions". Thesis, Lille 1, 2015. http://www.theses.fr/2015LIL10038/document.
Texto completoStrategic decision-making over valuable resources should consider risk-averse objectives. Many practical areas of application consider risk as central to decision-making. However, machine learning does not. As a result, research should provide insights and algorithms that endow machine learning with the ability to consider decision-theoretic risk. In particular, in estimating decision-theoretic risk on short dependent sequences generated from the most general possible class of processes for statistical inference and through decision-theoretic risk objectives in sequential decision-making. This thesis studies these two problems to provide principled algorithmic methods for considering decision-theoretic risk in machine learning. An algorithm with state-of-the-art performance is introduced for accurate estimation of risk statistics on the most general class of stationary--ergodic processes and risk-averse objectives are introduced in sequential decision-making (online learning) in both the stochastic multi-arm bandit setting and the adversarial full-information setting
Jalam, Radwan Chauchat Jean-Hugues. "Apprentissage automatique et catégorisation de textes multilingues". Lyon : Université Lumière Lyon 2, 2003. http://demeter.univ-lyon2.fr/sdx/theses/lyon2/2003/jalam_r.
Texto completoGarlet, Milani Luís Felipe. "Autotuning assisté par apprentissage automatique de tâches OpenMP". Thesis, Université Grenoble Alpes, 2020. http://www.theses.fr/2020GRALM022.
Texto completoModern computer architectures are highly complex, requiring great programming effort to obtain all the performance the hardware is capable of delivering. Indeed, while developers know potential optimizations, the only feasible way to tell which of them is faster for some platform is to test it. Furthermore, the many differences between two computer platforms, in the number of cores, cache sizes, interconnect, processor and memory frequencies, etc, makes it very challenging to have the same code perform well over several systems. To extract the most performance, it is often necessary to fine-tune the code for each system. Consequently, developers adopt autotuning to achieve some degree of portable performance. This way, the potential optimizations can be specified once, and, after testing each possibility on a platform, obtain a high-performance version of the code for that particular platform. However, this technique requires tuning each application for each platform it targets. This is not only time consuming but the autotuning and the real execution of the application differ. Differences in the data may trigger different behaviour, or there may be different interactions between the threads in the autotuning and the actual execution. This can lead to suboptimal decisions if the autotuner chooses a version that is optimal for the training but not for the real execution of the application. We propose the use of autotuning for selecting versions of the code relevant for a range of platforms and, during the execution of the application, the runtime system identifies the best version to use using one of three policies we propose: Mean, Upper Confidence Bound, and Gradient Bandit. This way, training effort is decreased and it enables the use of the same set of versions with different platforms without sacrificing performance. We conclude that the proposed policies can identify the version to use without incurring substantial performance losses. Furthermore, when the user does not know enough details of the application to configure optimally the explore-then-commit policy usedy by other runtime systems, the more adaptable UCB policy can be used in its place
Do, Quoc khanh. "Apprentissage discriminant des modèles continus en traduction automatique". Thesis, Université Paris-Saclay (ComUE), 2016. http://www.theses.fr/2016SACLS071/document.
Texto completoOver the past few years, neural network (NN) architectures have been successfully applied to many Natural Language Processing (NLP) applications, such as Automatic Speech Recognition (ASR) and Statistical Machine Translation (SMT).For the language modeling task, these models consider linguistic units (i.e words and phrases) through their projections into a continuous (multi-dimensional) space, and the estimated distribution is a function of these projections. Also qualified continuous-space models (CSMs), their peculiarity hence lies in this exploitation of a continuous representation that can be seen as an attempt to address the sparsity issue of the conventional discrete models. In the context of SMT, these echniques have been applied on neural network-based language models (NNLMs) included in SMT systems, and oncontinuous-space translation models (CSTMs). These models have led to significant and consistent gains in the SMT performance, but are also considered as very expensive in training and inference, especially for systems involving large vocabularies. To overcome this issue, Structured Output Layer (SOUL) and Noise Contrastive Estimation (NCE) have been proposed; the former modifies the standard structure on vocabulary words, while the latter approximates the maximum-likelihood estimation (MLE) by a sampling method. All these approaches share the same estimation criterion which is the MLE ; however using this procedure results in an inconsistency between theobjective function defined for parameter stimation and the way models are used in the SMT application. The work presented in this dissertation aims to design new performance-oriented and global training procedures for CSMs to overcome these issues. The main contributions lie in the investigation and evaluation of efficient training methods for (large-vocabulary) CSMs which aim~:(a) to reduce the total training cost, and (b) to improve the efficiency of these models when used within the SMT application. On the one hand, the training and inference cost can be reduced (using the SOUL structure or the NCE algorithm), or by reducing the number of iterations via a faster convergence. This thesis provides an empirical analysis of these solutions on different large-scale SMT tasks. On the other hand, we propose a discriminative training framework which optimizes the performance of the whole system containing the CSM as a component model. The experimental results show that this framework is efficient to both train and adapt CSM within SMT systems, opening promising research perspectives
Fradet, Nathan. "Apprentissage automatique pour la modélisation de musique symbolique". Electronic Thesis or Diss., Sorbonne université, 2024. https://accesdistant.sorbonne-universite.fr/login?url=https://theses-intra.sorbonne-universite.fr/2024SORUS037.pdf.
Texto completoSymbolic music modeling (SMM) represents the tasks performed by Deep Learning models on the symbolic music modality, among which are music generation or music information retrieval. SMM is often handled with sequential models that process data as sequences of discrete elements called tokens. This thesis study how symbolic music can be tokenized, and what are the impacts of the different ways to do it impact models performances and efficiency. Current challenges include the lack of software to perform this step, poor model efficiency and inexpressive tokens. We address these challenges by: 1) developing a complete, flexible and easy to use software library allowing to tokenize symbolic music; 2) analyzing the impact of various tokenization strategies on model performances; 3) increasing the performance and efficiency of models by leveraging large music vocabularies with the use of byte pair encoding; 4) building the first large-scale model for symbolic music generation
Goix, Nicolas. "Apprentissage automatique et extrêmes pour la détection d'anomalies". Thesis, Paris, ENST, 2016. http://www.theses.fr/2016ENST0072/document.
Texto completoAnomaly detection is not only a useful preprocessing step for training machine learning algorithms. It is also a crucial component of many real-world applications, from various fields like finance, insurance, telecommunication, computational biology, health or environmental sciences. Anomaly detection is also more and more relevant in the modern world, as an increasing number of autonomous systems need to be monitored and diagnosed. Important research areas in anomaly detection include the design of efficient algorithms and their theoretical study but also the evaluation of such algorithms, in particular when no labeled data is available -- as in lots of industrial setups. In other words, model design and study, and model selection. In this thesis, we focus on both of these aspects. We first propose a criterion for measuring the performance of any anomaly detection algorithm. Then we focus on extreme regions, which are of particular interest in anomaly detection, to obtain lower false alarm rates. Eventually, two heuristic methods are proposed, the first one to evaluate anomaly detection algorithms in the case of high dimensional data, the other to extend the use of random forests to the one-class setting
Goix, Nicolas. "Apprentissage automatique et extrêmes pour la détection d'anomalies". Electronic Thesis or Diss., Paris, ENST, 2016. http://www.theses.fr/2016ENST0072.
Texto completoAnomaly detection is not only a useful preprocessing step for training machine learning algorithms. It is also a crucial component of many real-world applications, from various fields like finance, insurance, telecommunication, computational biology, health or environmental sciences. Anomaly detection is also more and more relevant in the modern world, as an increasing number of autonomous systems need to be monitored and diagnosed. Important research areas in anomaly detection include the design of efficient algorithms and their theoretical study but also the evaluation of such algorithms, in particular when no labeled data is available -- as in lots of industrial setups. In other words, model design and study, and model selection. In this thesis, we focus on both of these aspects. We first propose a criterion for measuring the performance of any anomaly detection algorithm. Then we focus on extreme regions, which are of particular interest in anomaly detection, to obtain lower false alarm rates. Eventually, two heuristic methods are proposed, the first one to evaluate anomaly detection algorithms in the case of high dimensional data, the other to extend the use of random forests to the one-class setting
Guérif, Sébastien. "Réduction de dimension en apprentissage numérique non supervisé". Paris 13, 2006. http://www.theses.fr/2006PA132032.
Texto completoManago, Michel. "Intégration de techniques numériques et symboliques en apprentissage automatique". Paris 11, 1988. http://www.theses.fr/1988PA112386.
Texto completoPerreault, Samuel. "Structures de corrélation partiellement échangeables : inférence et apprentissage automatique". Doctoral thesis, Université Laval, 2020. http://hdl.handle.net/20.500.11794/66443.
Texto completoHANSER, THIERRY. "Apprentissage automatique de methodes de synthese a partir d'exemples". Université Louis Pasteur (Strasbourg) (1971-2008), 1993. http://www.theses.fr/1993STR13106.
Texto completoCoulibaly, Lassana. "Contribution en apprentissage automatique pour la maîtrise des risques". Thesis, Toulouse, INPT, 2020. http://www.theses.fr/2020INPT0109.
Texto completoClimate change regularly causes phenomena that directly threaten the environment and humanity. In this context, meteorology is playing more and more an important role in the understanding and forecasting of these phenomena. The problems of reliability of the observations is essential for the numerical reasoning and the quality of the simulation. In addition, interoperability is important both for companies and for public services dealing with complex data and models. In meteorological services, the reliability of observational data is a fundamental requirement. Weather and climate predictions are dependent on many physical phenomena on different time and space scales. One of these phenomena is the transfer of energy from the surface to the atmosphere that is a sensitive parameter. Observations of sensitive parameters often produce data that are unreliable (imperfect data). A better treatment of these imperfect data may improve the evaluation of the simulation. We propose the use of machine learning methods that can : (i) improve the evaluation of surface-atmosphere exchanges in numerical weather and climate prediction models and (ii) produce knowledge for interoperability. This can support the communication of observation services and numerical prediction models. The objective of this work is to diagnose numerical prediction models in order to look for the weaknesses of these models in the simulation of exchanges between the surface and the atmosphere. These exchanges are quantified by sensible and latent heat fluxes. In a first instance, Gaussian processes taking into account uncertainties are used to model the measured values in order to make the observational database more reliable. This modelling is carried out through a thorough learning process that includes regression by integrating field knowledge. Then the extraction method of the association rules is chosen in order to : highlight the weaknesses of the model ; make comparisons between the observations made and the simulations made by the numerical model. Finally, an optimizer has been defined from some properties on geometric transformations in mathematics. This optimizer makes it possible to perform an adjustment to the simulated data in order to minimize simulation errors. These methods are deployed on a measured data base on the experimental site of the Centre de Recherches Atmosphériques (CRA) which is one of the two sites making up the Pyrénéenne Plateforme d'Observation de l'Atmosphère (P2OA) in France. The results obtained and expressed in the form of association rules have made it possible to highlight certain weaknesses in the numerical models : first, the highlighting of differences (errors) between the observations and the simulations ; then the analysis of the generated rules showed that important differences on global radiation are often concomitant with important differences on sensible and latent heat fluxes. This is often due to natural disturbances (e.g. cloud location) that impact the quality of observations/simulations of sensible and latent heat fluxes. The expected benefits are related to the generation of useful knowledge to improve the quality of numerical simulation of surface processes. In addition, the proposed optimizer gave satisfactory results. The simulated values were scaled to 100% in the case of similar shapes and to 98% in the case of shapes with peaks. This optimizer can be applied to all other meteorological variables
Margeta, Ján. "Apprentissage automatique pour simplifier l’utilisation de banques d’images cardiaques". Thesis, Paris, ENMP, 2015. http://www.theses.fr/2015ENMP0055/document.
Texto completoThe recent growth of data in cardiac databases has been phenomenal. Cleveruse of these databases could help find supporting evidence for better diagnosis and treatment planning. In addition to the challenges inherent to the large quantity of data, the databases are difficult to use in their current state. Data coming from multiple sources are often unstructured, the image content is variable and the metadata are not standardised. The objective of this thesis is therefore to simplify the use of large databases for cardiology specialists withautomated image processing, analysis and interpretation tools. The proposed tools are largely based on supervised machine learning techniques, i.e. algorithms which can learn from large quantities of cardiac images with groundtruth annotations and which automatically find the best representations. First, the inconsistent metadata are cleaned, interpretation and visualisation of images is improved by automatically recognising commonly used cardiac magnetic resonance imaging views from image content. The method is based on decision forests and convolutional neural networks trained on a large image dataset. Second, the thesis explores ways to use machine learning for extraction of relevant clinical measures (e.g. volumes and masses) from3D and 3D+t cardiac images. New spatio-temporal image features are designed andclassification forests are trained to learn how to automatically segment the main cardiac structures (left ventricle and left atrium) from voxel-wise label maps. Third, a web interface is designed to collect pairwise image comparisons and to learn how to describe the hearts with semantic attributes (e.g. dilation, kineticity). In the last part of the thesis, a forest-based machinelearning technique is used to map cardiac images to establish distances and neighborhoods between images. One application is retrieval of the most similar images
Manago, Michel. "Intégration de techniques numériques et symboliques en apprentissage automatique". Grenoble 2 : ANRT, 1988. http://catalogue.bnf.fr/ark:/12148/cb37619406x.
Texto completoCrochepierre, Laure. "Apprentissage automatique interactif pour les opérateurs du réseau électrique". Electronic Thesis or Diss., Université de Lorraine, 2022. http://www.theses.fr/2022LORR0112.
Texto completoIn the energy transition context and the increase in interconnections between the electricity transmission networks in Europe, the French network operators must now deal with more fluctuations and new network dynamics. To guarantee the safety of the network, operators rely on computer software that allows them to carry out simulations or to monitor the evolution of indicators created manually by experts, thanks to their knowledge of the operation of the network. The French electricity transmission network operator RTE (Réseau de Transport d'Electricité) is particularly interested in developing tools to assist operators in monitoring flows on power lines. Flows are notably important to maintain the network in a safe state, guaranteeing the safety of equipment and people. However, the indicators used are not easy to update because of the expertise required to construct and analyze them.In order to address the stated problem, this thesis aims at constructing indicators, in the form of symbolic expressions, to estimate flows on power lines. The problem is studied from the Symbolic Regression perspective and investigated using both Grammatical Evolution and Reinforcement Learning approaches in which explicit and implicit expert knowledge is taken into account. Explicit knowledge about the physics and expertise of the electrical domain is represented in the form of a Context-Free Grammar to limit the functional space from which an expression is created. A first approach of Interactive Grammatical Evolution proposes to incrementally improve found expressions by updating a grammar between evolutionary learnings. Expressions are obtained on real-world data from the network history, validated by an analysis of learning metrics and an interpretability evaluation. Secondly, we propose a reinforcement approach to search in a space delimited by a Context-Free Grammar in order to build a relevant symbolic expression to applications involving physical constraints. This method is validated on state-of-the-art Symbolic Regression benchmarks and also on a dataset with physical constraints to assess its interpretability.Furthermore, in order to take advantage of the complementarities between the capacities of machine learning algorithms and the expertise of network operators, interactive Symbolic Regression algorithms are proposed and integrated into interactive platforms. Interactivity allows updating the knowledge represented in grammatical form and analyzing, interacting with, and commenting on the solutions found by the different approaches. These algorithms and interactive interfaces also aim to take into account implicit knowledge, which is more difficult to formalize, through interaction mechanisms based on suggestions and user preferences
Jacques, Céline. "Méthodes d'apprentissage automatique pour la transcription automatique de la batterie". Electronic Thesis or Diss., Sorbonne université, 2019. http://www.theses.fr/2019SORUS150.
Texto completoThis thesis focuses on learning methods for automatic transcription of the battery. They are based on a transcription algorithm using a non-negative decomposition method, NMD. This thesis raises two main issues: the adaptation of methods to the analyzed signal and the use of deep learning. Taking into account the information of the signal analyzed in the model can be achieved by their introduction during the decomposition steps. A first approach is to reformulate the decomposition step in a probabilistic context to facilitate the introduction of a posteriori information with methods such as SI-PLCA and statistical NMD. A second approach is to implement an adaptation strategy directly in the NMD: the application of modelable filters to the patterns to model the recording conditions or the adaptation of the learned patterns directly to the signal by applying strong constraints to preserve their physical meaning. The second approach concerns the selection of the signal segments to be analyzed. It is best to analyze segments where at least one percussive event occurs. An onset detector based on a convolutional neural network (CNN) is adapted to detect only percussive onsets. The results obtained being very interesting, the detector is trained to detect only one instrument allowing the transcription of the three main drum instruments with three CNNs. Finally, the use of a CNN multi-output is studied to transcribe the part of battery with a single network
Njike, Fotzo Hermine. "Structuration Automatique de Corpus Textuels par Apprentissage Automatique : Automatically structuring textual corpora with machine learning methods". Paris 6, 2004. http://www.theses.fr/2004PA066567.
Texto completoDuclaye, Florence. "Apprentissage automatique de relations d'équivalence sémantique à partir du Web". Phd thesis, Télécom ParisTech, 2003. http://pastel.archives-ouvertes.fr/pastel-00001119.
Texto completoBernhard, Delphine. "Apprentissage de connaissances morphologiques pour l'acquisition automatique de ressources lexicales". Phd thesis, Université Joseph Fourier (Grenoble), 2006. http://tel.archives-ouvertes.fr/tel-00119257.
Texto completoNous présentons deux systèmes d'acquisition de connaissances morphologiques non supervisés, caractérisés par des approches différentes. Le premier procède par segmentation des mots, tandis que le second regroupe les mots dans des familles morphologiques.
Nous explorons ensuite les utilisations possibles de ce type d'informations pour l'acquisition de termes et de relations sémantiques. Nous proposons notamment une méthode de pondération et de visualisation des mots clés extraits de corpus de textes de spécialité en fonction de leur famille morphologique. Nous définissons également des schémas, basés sur les résultats de la segmentation morphologique, afin de découvrir des relations sémantiques telles que la spécialisation et la cohyponymie.
Nicolas, Jacques. "Ally, un systeme logique pour la generalisation en apprentissage automatique". Rennes 1, 1987. http://www.theses.fr/1987REN10043.
Texto completoDuclaye, Florence Aude Dorothée. "Apprentissage automatique de relations d'équivalence sémantique à partir du Web". Paris, ENST, 2003. http://www.theses.fr/2003ENST0044.
Texto completoThis PhD thesis can be situated in the context of a question answering system, which is capable of automatically finding answers to factual questions on the Web. One way to improve the quality of these answers is to increase the recall rate of the system, by identifying the answers under multiple possible formulations(paraphrases). As the manual recording of paraphrases is a long and expensive task, the goal of this PhD thesis is to design and develop a mechanism that learns automatically and in a weakly supervised manner the possible paraphrases of an answer. Thanks to the redundance and the linguistic variety of the information it contains, the Web is considered to be a very interesting corpus. Assimilated to a gigantic bipartite graph represented, on the one hand, by formulations and, on the other hand, by argument couples, the Web turns out to be propitious to the application of Firth's hypothesis, according to which "you shall know a word (resp. A formulation, in our case) by the company (resp. Arguments) it keeps". Consequently, the Web is sampled using an iterative mechanism : formulations (potential paraphrases) are extracted by anchoring arguments and, inversely, new arguments are extracted by anchoring the acquired formulations. In order to make the learning process converge, an intermediary stage is necessary, which partitions the sampled data using a statistical classification method. The obtained results were empirically evaluated, which, more particularly, shows the value added by the learnt paraphrases of the question answering system
Morlec, Yann. "Génération multiparamétrique de la prosodie du français par apprentissage automatique". Grenoble INPG, 1997. http://www.theses.fr/1997INPG0221.
Texto completoFerreira, Emmanuel. "Apprentissage automatique en ligne pour un dialogue homme-machine situé". Thesis, Avignon, 2015. http://www.theses.fr/2015AVIG0206/document.
Texto completoA dialogue system should give the machine the ability to interactnaturally and efficiently with humans. In this thesis, we focus on theissue of the development of stochastic dialogue systems. Thus, we especiallyconsider the Partially Observable Markov Decision Process (POMDP)framework which yields state-of-the-art performance on goal-oriented dialoguemanagement tasks. This model enables the system to cope with thecommunication ambiguities due to noisy channel and also to optimize itsdialogue management strategy directly from data with Reinforcement Learning (RL)methods.Considering statistical approaches often requires the availability of alarge amount of training data to reach good performance. However, corpora of interest are seldom readily available and collectingsuch data is both time consuming and expensive. For instance, it mayrequire a working prototype to initiate preliminary experiments with thesupport of expert users or to consider other alternatives such as usersimulation techniques.Very few studies to date have considered learning a dialogue strategyfrom scratch by interacting with real users, yet this solution is ofgreat interest. Indeed, considering the learning process as part of thelife cycle of a system offers a principle framework to dynamically adaptthe system to new conditions in an online and seamless fashion.In this thesis, we endeavour to provide solutions to make possible thisdialogue system cold start (nearly from scratch) but also to improve its ability to adapt to new conditions in operation (domain extension, new user profile, etc.).First, we investigate the conditions under which initial expertknowledge (such as expert rules) can be used to accelerate the policyoptimization of a learning agent. Similarly, we study how polarized userappraisals gathered throughout the course of the interaction can beintegrated into a reinforcement learning-based dialogue manager. Morespecifically, we discuss how this information can be cast intosocially-inspired rewards to speed up the policy optimisation for bothefficient task completion and user adaptation in an online learning setting.The results obtained on a reference task demonstrate that a(quasi-)optimal policy can be learnt in just a few hundred dialogues,but also that the considered additional information is able tosignificantly accelerate the learning as well as improving the noise tolerance.Second, we focus on reducing the development cost of the spoken language understanding module. For this, we exploit recent word embedding models(projection of words in a continuous vector space representing syntacticand semantic properties) to generalize from a limited initial knowledgeabout the dialogue task to enable the machine to instantly understandthe user utterances. We also propose to dynamically enrich thisknowledge with both active learning techniques and state-of-the-artstatistical methods. Our experimental results show that state-of-the-artperformance can be obtained with a very limited amount of in-domain andin-context data. We also show that we are able to refine the proposedmodel by exploiting user returns about the system outputs as well as tooptimize our adaptive learning with an adversarial bandit algorithm tosuccessfully balance the trade-off between user effort and moduleperformance.Finally, we study how the physical embodiment of a dialogue system in a humanoid robot can help the interaction in a dedicated Human-Robotapplication where dialogue system learning and testing are carried outwith real users. Indeed, in this thesis we propose an extension of thepreviously considered decision-making techniques to be able to take intoaccount the robot's awareness of the users' belief (perspective taking)in a RL-based situated dialogue management optimisation procedure
Nicolas, Jacques. "ALLY, un système logique pour la généralisation en apprentissage automatique". Grenoble 2 : ANRT, 1987. http://catalogue.bnf.fr/ark:/12148/cb37608434q.
Texto completoDuclaye, Florence. "Apprentissage automatique de relations d'équivalence sémantique à partir du Web /". Paris : École nationale supérieure des télécommunications, 2005. http://catalogue.bnf.fr/ark:/12148/cb39935321s.
Texto completoBen, Hassena Anouar. "Apprentissage par analogie de structures d'arbres". Rennes 1, 2012. http://www.theses.fr/2011REN1E007.
Texto completoTeytaud, Olivier. "Apprentissage, réseaux de neurones et applications". Lyon 2, 2001. http://theses.univ-lyon2.fr/documents/lyon2/2001/teytaud_o.
Texto completoTeytaud, Olivier Paugam-Moisy Hélène. "Apprentissage, réseaux de neurones et applications". [S.l.] : [s.n.], 2001. http://demeter.univ-lyon2.fr:8080/sdx/theses/lyon2/2001/teytaud_o.
Texto completoLefort, Riwal. "Apprentissage et classification faiblement supervisée : application en acoustique halieutique". Télécom Bretagne, 2010. http://www.theses.fr/2010TELB0164.
Texto completoThis thesis deals with object classification and weakly supervised learning. An application to fisheries acoustics is considered. In weakly supervised learning, the training dataset is weakly annotated, i. E. The class knowledge is given by prior. Formally, each training object is associated with vector that provides the prior for each class. In this context, we investigate generative model, discriminative model and a model based on random forest. Furthermore, an iterative procedure is proposed for modifying uncertain priors from low value to more certain value. Experiments are carried out to evaluate classification models as regards to prior complexity. In order to control the prior complexity, weakly supervised dataset are generated from supervised dataset. In fisheries acoustics, fish schools in images are classified, the objective being to study an ecosystem or to assess fish stock biomass. Fish species identification is carried out by trawl catches that provide species prior in a given area. In this context, the new multibeam echosunder provides 3D images. These images are richer and more informative than 2D monobeam echosounder. Firstly, we propose a new global descriptor for characterizing fish school images. The descriptor models both the spatial distribution of fish schools in images and the type of fish schools. Secondly, we propose to apply weakly supervised training schemes to assess fish school biomass in the Bay of Biscay
Charton, Eric. "Génération de phrases multilingues par apprentissage automatique de modèles de phrases". Phd thesis, Université d'Avignon, 2010. http://tel.archives-ouvertes.fr/tel-00622561.
Texto completoMurgue, Thierry. "Extraction de données et apprentissage automatique pour les sites web adaptatifs". Phd thesis, Ecole Nationale Supérieure des Mines de Saint-Etienne, 2006. http://tel.archives-ouvertes.fr/tel-00366586.
Texto completoCharton, Éric. "Génération de phrases multilingues par apprentissage automatique de modèles de phrases". Thesis, Avignon, 2010. http://www.theses.fr/2010AVIG0175/document.
Texto completoNatural Language Generation (NLG) is the natural language processing task of generating natural language from a machine representation system. In this thesis report, we present an architecture of NLG system relying on statistical methods. The originality of our proposition is its ability to use a corpus as a learning resource for sentences production. This method offers several advantages : it simplifies the implementation and design of a multilingual NLG system, capable of sentence production of the same meaning in several languages. Our method also improves the adaptability of a NLG system to a particular semantic field. In our proposal, sentence generation is achieved trough the use of sentence models, obtained from a training corpus. Extracted sentences are abstracted by a labelling step obtained from various information extraction and text mining methods like named entity recognition, co-reference resolution, semantic labelling and part of speech tagging. The sentence generation process is achieved by a sentence realisation module. This module provide an adapted sentence model to fit a communicative intent, and then transform this model to generate a new sentence. Two methods are proposed to transform a sentence model into a generated sentence, according to the semantic content to express. In this document, we describe the complete labelling system applied to encyclopaedic content to obtain the sentence models. Then we present two models of sentence generation. The first generation model substitute the semantic content to an original sentence content. The second model is used to find numerous proto-sentences, structured as Subject, Verb, Object, able to fit by part a whole communicative intent, and then aggregate all the selected proto-sentences into a more complex one. Our experiments of sentence generation with various configurations of our system have shown that this new approach of NLG have an interesting potential
Caillaud, Bertrand. "Apprentissage de connaissances prosodiques pour la reconnaissance automatique de la parole". Grenoble INPG, 1996. http://www.theses.fr/1996INPG0219.
Texto completoOuld, Abdel Vetah Mohamed. "Apprentissage automatique appliqué à l'extraction d'information à partir de textes biologiques". Paris 11, 2005. http://www.theses.fr/2005PA112133.
Texto completoThis thesis is about information extraction from textual data. Two main approaches co-exist in this field. The first approach is based on shallow text analysis. These methods are easy to implement but the information they extract is often incomplete and noisy. The second approach requires deeper structural linguistic information. Compared to the first approach, it has the double advantage of being easily adaptable and of taking into account the diversity of formulation which is an intrinsic characteristic of textual data. In this thesis, we have contributed to the realization of a complete information extraction tool based on this latter approach. Our tool is dedicated to the automatic extraction of gene interactions described in MedLine abstracts. In the first part of the work, we develop a filtering module that allows the user to identify the sentences referring to gene interactions. The module is available on line and already used by biologists. The second part of the work introduces an original methodology based on an abstraction of the syntactic analysis for automatical learning of information extraction rules. The preliminary results are promising and show that our abstraction approach provides a good representation for learning extraction rules
Vu, Viet-Vu. "Clustering semi-supervisé et apprentissage actif". Paris 6, 2011. http://www.theses.fr/2011PA066607.
Texto completoBoyer, Laurent. "Apprentissage probabiliste de similarités d'édition". Phd thesis, Université Jean Monnet - Saint-Etienne, 2011. http://tel.archives-ouvertes.fr/tel-00718835.
Texto completoPessiot, Jean-François. "Apprentissage automatique pour l'extraction de caractéristiques : application au partitionnement de documents, au résumé automatique et au filtrage collaboratif". Paris 6, 2008. http://www.theses.fr/2008PA066218.
Texto completoLe, Lann Marie-Véronique. "Commande prédictive et commande par apprentissage : étude d'une unité pilote d'extraction, optimisation par apprentissage". Toulouse, INPT, 1988. http://www.theses.fr/1988INPT023G.
Texto completoBurg, Bernard. "Apprentissage de règles de comportement destinées au contrôle d'un système". Paris 11, 1988. http://www.theses.fr/1988PA112375.
Texto completoProcess control systems have to face applications which are always more ambitions and difficult to master. In some cases it is not easy to use conventional process control techniques. With the introduction of declarative methods it is possible to start in a pragmatic way and to set an implicit formulation of the problem when no explicit formulation is available. New mechanisms can be envisioned, and we conceived a rule based controller, then the difficulty remains on the design of the rule sets. To overcome this problem, we had to use jointly some learning techniques, such as data analysis to cope with noisy data and to project them into reduced space representations. Then structural techniques allow to modelise the temporal evolution of the process control and the hidden structures. Finally, artificial intelligence machine learning techniques discover the concepts and generalise the acquired knowledge. The whole technique set is supervised by artificial intelligence, it analyses the results issued from each learning step and planes the next action to perform. Three learning strategies are used: the first one starts from the data and uses inductive learning, it proves some completeness. The second one begins with a fuzzy model and acquires rules by deduction, it brings coherency via expert knowledge. Finally the behavior rules are used and refined by means of interaction with the environment. The learning program CANDIDE performed two case studies - the speed control of a DC motor the automatic driving of a car
Lecerf, Loïc. "L' apprentissage machine pour assister l'annotation de documents : clustering visuel interactif, apprentissage actif et extraction automatique des descripteurs". Paris 6, 2009. http://www.theses.fr/2009PA066186.
Texto completo