Dissertations / Theses on the topic '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 '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.
Roussel, Jacques. "Evaluation et apprentissage : l'evaluation en arithmetique au cycle des apprentissages." Toulouse 2, 1989. http://www.theses.fr/1989TOU20062.
Full textTeachers hope for a theory of evaluation that precsibes means and conditions of correct evaluation. Epistemological treatment of didactic evaluationshows that all standard conception entail the destruction of its object: relativism, differenciation and diversity are the essential properties of the treatment of pupils' answers. That view is applied to evaluation in arithmetic in the beginners'classes. The hypothesis of "answering evaluation", linked to the "answering teching", a theory developed by l. Not, explains two facts: the preservation of the propose of didactic evaluation, i. E. The appropriation of knowledge, its structure and genesis; the protection against a cheating operator that pretends to turn an answer into a result. When the conditions of answering evaluation are not observed, evaluation becomes a pseudo or a meta-evaluation. The proof is established after a critical review of recent papers about the didactics of arithmetic and an analysis of a variety of non didactic evaluation. The purpose of the thesis was to examine two possibilities : the practice of evaluation by the teacher in the class-room and the content of the training of teachers for evaluation
Venturini, Gilles. "Apprentissage adaptatif et apprentissage supervise par algorithme genetique." Paris 11, 1994. http://www.theses.fr/1994PA112016.
Full textMoradi, Fard Maziar. "Apprentissage de représentations de données dans un apprentissage non-supervisé." Thesis, Université Grenoble Alpes, 2020. http://www.theses.fr/2020GRALM053.
Full textDue to the great impact of deep learning on variety fields of machine learning, recently their abilities to improve clustering approaches have been investi- gated. At first, deep learning approaches (mostly Autoencoders) have been used to reduce the dimensionality of the original space and to remove possible noises (also to learn new data representations). Such clustering approaches that utilize deep learning approaches are called Deep Clustering. This thesis focuses on developing Deep Clustering models which can be used for different types of data (e.g., images, text). First we propose a Deep k-means (DKM) algorithm where learning data representations (through a deep Autoencoder) and cluster representatives (through the k-means) are performed in a joint way. The results of our DKM approach indicate that this framework is able to outperform similar algorithms in Deep Clustering. Indeed, our proposed framework is able to truly and smoothly backpropagate the loss function error through all learnable variables.Moreover, we propose two frameworks named SD2C and PCD2C which are able to integrate respectively seed words and pairwise constraints into end-to-end Deep Clustering frameworks. In fact, by utilizing such frameworks, the users can observe the reflection of their needs in clustering. Finally, the results obtained from these frameworks indicate their ability to obtain more tailored results
Boucheron, Stéphane. "Apprentissage et calculs." Montpellier 2, 1988. http://www.theses.fr/1988MON20251.
Full textBoucheron, Stéphane. "Apprentissage et calculs." Grenoble 2 : ANRT, 1988. http://catalogue.bnf.fr/ark:/12148/cb376121172.
Full textSalomon, Antoine. "Apprentissage stratégique statistique." Paris 13, 2010. http://www.theses.fr/2010PA132039.
Full textThis thesis studies strategic interaction between several agents who are facing an exploration vs. Exploitation dilemma. In game theory, this situation is well described by models of bandit games. Each player faces a two-arm bandit machine, one arm being safe, the other being risky. At each stage of the game, each player has to decide which arm he uses. If he chooses the risky arm (exploration), he gets a random payoff which gives him partial information on the rentability of his machine. If he chooses the safe arm, he gets a known payoff, but possibly less than what he could have got from exploration. The rentability of the machine depends on an unknown state of the nature, which can be learnt from exploration. Learning is a strategic issue: for instance a player could benefit from others' information without taking risks himself. We study Nash equilibria of such games. We mainly wonder if equilibria are efficient: does a player gain significanlty more from strategic interaction than he would alone? Is there some kind of cooperation that helps getting more information? Do players manage to have a good knowledge of the state of the nature? This depends on what agents are able to see from each other (actions and/or payoffs), and also on how the types of the machines are correlated. We will also study the way equilibria are evolving when the number of players get large. In particular, we wonder if this increase leads to better pieces of information, and better gains
Dinh, Quang-Thang. "Apprentissage statistique relationnel : apprentissage de structures de réseaux de Markov logiques." Phd thesis, Université d'Orléans, 2011. http://tel.archives-ouvertes.fr/tel-00659738.
Full textLaronze, Florian. "Apprentissage à distance, apprentissage gamifié : identification des facteurs de la réussite universitaire." Electronic Thesis or Diss., Bordeaux, 2024. http://www.theses.fr/2024BORD0445.
Full textRecently, the President of France Université emphasised that ‘the ultimate objective of the French university is to integrate those who want to enrol and to ensure that they succeed’. Certainly exacerbated by the major health crisis caused by the Covid-19 pandemic, a high proportion of students are experiencing serious difficulties, with significant repercussions for their mental health. In addition, just over a quarter of students enrolled on a Licence degree complete their three-year programme. Given these facts, the need for a better understanding of the factors that determine success at university is a major challenge for our society. Also associated with the health crisis, the practice of distance learning, and in particular its synchronous mode (e.g. Zoom), has boomed worldwide and seems destined to become a common mode of learning. Coupled with this development of distance learning, but also with the technological revolution and pedagogical considerations aimed at encouraging student engagement during lessons, recent years have also seen the rise of digital and gamified teaching tools (e.g. mobile quiz applications, virtual reality environments for educational purposes, etc.). In this general context, the aim of my thesis is to gain a better understanding of the impact of distance learning and gamified digital tools on academic success (e.g. emotions felt, motivation, grades, etc.). As part of the national ‘Université Atypie Friendly’ project, which aims to promote the academic inclusion of young adults with Autism Spectrum Disorder, this thesis will also seek to identify the most relevant digital tools and pedagogical practices for this specific audience
Solnon, Matthieu. "Apprentissage statistique multi-tâches." Phd thesis, Université Pierre et Marie Curie - Paris VI, 2013. http://tel.archives-ouvertes.fr/tel-00911498.
Full textKocák, Tomáš. "Apprentissage séquentiel avec similitudes." Thesis, Lille 1, 2016. http://www.theses.fr/2016LIL10230/document.
Full textThis thesis studies several extensions of multi-armed bandit problem, where a learner sequentially selects an action and obtain the reward of the action. Traditionally, the only information the learner acquire is about the obtained reward while information about other actions is hidden from the learner. This limited feedback can be restrictive in some applications like recommender systems, internet advertising, packet routing, etc. Usually, these problems come with structure, similarities between users or actions, additional observations, or any additional assumptions. Therefore, it is natural to incorporate these assumptions to the algorithms to improve their performance. This thesis focuses on multi-armed bandit problem with some underlying structure usually represented by a graph with actions as vertices. First, we study a problem where the graph captures similarities between actions; connected actions tend to grand similar rewards. Second, we study a problem where the learner observes rewards of all the neighbors of the selected action. We study these problems under several additional assumptions on rewards (stochastic, adversarial), side observations (adversarial, stochastic, noisy), actions (one node at the time, several nodes forming a combinatorial structure in the graph). The main contribution of this thesis is to design algorithms for previously mentioned problems together with theoretical and empirical guaranties. We also introduce several novel quantities, to capture the difficulty of some problems, like effective dimension and effective independence number
Guritanu, Elena. "Types d'écriture et apprentissage." Thesis, Sorbonne Paris Cité, 2016. http://www.theses.fr/2016USPCB217/document.
Full textBefore being more specifically dedicated to the learning of writing, this thesis concentrates on an important part of the genesis and evolution of writing. It analyses the diverse states and forms of writing systems known since the great civilisations that originated them to our modern societies, and exposes various problems which form what we call here "the field of writing" : the invention of writing, its beginnings, its evolution in the myths and legends, the connections between the image and the language, the mechanics of this evolution, the questions of society that the written word poses to its treatment and the epistemological linguistic theories that it relates to. The examination of this field, particularly throws light on the second part of this study articulated around questions about the technicalities of writing. Based on a corpus made of five graphic systems - Chinese idiographic, Arabic consonantal writing and the Russian , Romanian and French alphabet systems, this study analyses and compares the teaching methods of reading and writing for each of them and supports the different learnings of writing in each system but also the important convergences shared from one system to an other
Scornet, Erwan. "Apprentissage et forêts aléatoires." Thesis, Paris 6, 2015. http://www.theses.fr/2015PA066533/document.
Full textThis is devoted to a nonparametric estimation method called random forests, introduced by Breiman in 2001. Extensively used in a variety of areas, random forests exhibit good empirical performance and can handle massive data sets. However, the mathematical forces driving the algorithm remain largely unknown. After reviewing theoretical literature, we focus on the link between infinite forests (theoretically analyzed) and finite forests (used in practice) aiming at narrowing the gap between theory and practice. In particular, we propose a way to select the number of trees such that the errors of finite and infinite forests are similar. On the other hand, we study quantile forests, a type of algorithms close in spirit to Breiman's forests. In this context, we prove the benefit of trees aggregation: while each tree of quantile forest is not consistent, with a proper subsampling step, the forest is. Next, we show the connection between forests and some particular kernel estimates, which can be made explicit in some cases. We also establish upper bounds on the rate of convergence for these kernel estimates. Then we demonstrate two theorems on the consistency of both pruned and unpruned Breiman forests. We stress the importance of subsampling to demonstrate the consistency of the unpruned Breiman's forests. At last, we present the results of a Dreamchallenge whose goal was to predict the toxicity of several compounds for several patients based on their genetic profile
Hamdi, Fatma. "Apprentissage en distributions déséquilibrées." Paris 13, 2012. http://scbd-sto.univ-paris13.fr/intranet/edgalilee_th_2012_hamdi.pdf.
Full textThe research work exposed in this thesis concerns the development of approaches for processing and modeling unbalanced databases. In order to afford solutions to this problem, we propose different contributions. A first proposition acting at the learning data level SNCR, it is a technique of adaptive structural sampling that allowing data rebalancing by sub-sampling of the ma jority class. The proposed method is guided by the topological structure of the data and their distribution. The second proposed approach in this thesis discuss the problem of one class learning, it is a way allowing to bypass the problem of unbalanced classes to a novelty detection problem. The model RS-NDF is based on a set of adaptive filters. Every filter is conceived in a description subspace which the components and the dimension are randomly chosen. Besides, we propose an improvement of the quality of RS-NFD by an extension SRS-NDF allowing to reduce the number of models participating in the decision. The goal is to choose between those filters the sub-set which allows to reach the best performances. Finally, we propose an adaptation of the RS-NDF approach to the concept drift detection problem. The results obtained using the proposed approaches are encouraging and promising
Magaud, François-Xavier. "Stratégies de méta-apprentissage." Lyon 1, 2007. http://www.theses.fr/2007LYO10081.
Full textThis study enables to build a meta-learning theory. It defines the meta-learning via a goal: to produce a meta-knowledge automatically. The meta-knowledge is generated by a learning strategy, called meta-learning. The theoretical framework is fixed by a classification in four types of meta-learning strategies : the selection type, the aggregation type, the self-adapting type and the reflexive type. Each type contains theoretical foundations, which is based on existing algorithms or on algorithms carried out for this thesis. An original tool for trajectories analysis was built to illustrate the reflexive type. It uses the meta-learning to define the meta-knowledge which has the shape of a curve
Bayoudh, Sabri. "Apprentissage par proportion analogique." Rennes 1, 2007. ftp://ftp.irisa.fr/techreports/theses/2007/bayoudh.pdf.
Full textThe 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
Zimmer, Matthieu. "Apprentissage par renforcement développemental." Thesis, Université de Lorraine, 2018. http://www.theses.fr/2018LORR0008/document.
Full textReinforcement learning allows an agent to learn a behavior that has never been previously defined by humans. The agent discovers the environment and the different consequences of its actions through its interaction: it learns from its own experience, without having pre-established knowledge of the goals or effects of its actions. This thesis tackles how deep learning can help reinforcement learning to handle continuous spaces and environments with many degrees of freedom in order to solve problems closer to reality. Indeed, neural networks have a good scalability and representativeness. They make possible to approximate functions on continuous spaces and allow a developmental approach, because they require little a priori knowledge on the domain. We seek to reduce the amount of necessary interaction of the agent to achieve acceptable behavior. To do so, we proposed the Neural Fitted Actor-Critic framework that defines several data efficient actor-critic algorithms. We examine how the agent can fully exploit the transitions generated by previous behaviors by integrating off-policy data into the proposed framework. Finally, we study how the agent can learn faster by taking advantage of the development of his body, in particular, by proceeding with a gradual increase in the dimensionality of its sensorimotor space
Zimmer, Matthieu. "Apprentissage par renforcement développemental." Electronic Thesis or Diss., Université de Lorraine, 2018. http://www.theses.fr/2018LORR0008.
Full textReinforcement learning allows an agent to learn a behavior that has never been previously defined by humans. The agent discovers the environment and the different consequences of its actions through its interaction: it learns from its own experience, without having pre-established knowledge of the goals or effects of its actions. This thesis tackles how deep learning can help reinforcement learning to handle continuous spaces and environments with many degrees of freedom in order to solve problems closer to reality. Indeed, neural networks have a good scalability and representativeness. They make possible to approximate functions on continuous spaces and allow a developmental approach, because they require little a priori knowledge on the domain. We seek to reduce the amount of necessary interaction of the agent to achieve acceptable behavior. To do so, we proposed the Neural Fitted Actor-Critic framework that defines several data efficient actor-critic algorithms. We examine how the agent can fully exploit the transitions generated by previous behaviors by integrating off-policy data into the proposed framework. Finally, we study how the agent can learn faster by taking advantage of the development of his body, in particular, by proceeding with a gradual increase in the dimensionality of its sensorimotor space
BONELLI, PIERRE. "Apprentissage par algorithmes genetiques." Paris 11, 1993. http://www.theses.fr/1993PA112395.
Full textLe, 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.
Full textLe, Lann Marie-Véronique. "Commande prédictive et commande par apprentissage étude d'une unité pilote d'extraction, optimisation par apprentissage /." Grenoble 2 : ANRT, 1988. http://catalogue.bnf.fr/ark:/12148/cb37615168p.
Full textBertrand, Hadrien. "Optimisation d'hyper-paramètres en apprentissage profond et apprentissage par transfert : applications en imagerie médicale." Thesis, Université Paris-Saclay (ComUE), 2019. http://www.theses.fr/2019SACLT001/document.
Full textIn the last few years, deep learning has changed irrevocably the field of computer vision. Faster, giving better results, and requiring a lower degree of expertise to use than traditional computer vision methods, deep learning has become ubiquitous in every imaging application. This includes medical imaging applications. At the beginning of this thesis, there was still a strong lack of tools and understanding of how to build efficient neural networks for specific tasks. Thus this thesis first focused on the topic of hyper-parameter optimization for deep neural networks, i.e. methods for automatically finding efficient neural networks on specific tasks. The thesis includes a comparison of different methods, a performance improvement of one of these methods, Bayesian optimization, and the proposal of a new method of hyper-parameter optimization by combining two existing methods: Bayesian optimization and Hyperband.From there, we used these methods for medical imaging applications such as the classification of field-of-view in MRI, and the segmentation of the kidney in 3D ultrasound images across two populations of patients. This last task required the development of a new transfer learning method based on the modification of the source network by adding new geometric and intensity transformation layers.Finally this thesis loops back to older computer vision methods, and we propose a new segmentation algorithm combining template deformation and deep learning. We show how to use a neural network to predict global and local transformations without requiring the ground-truth of these transformations. The method is validated on the task of kidney segmentation in 3D US images
Bertrand, Hadrien. "Optimisation d'hyper-paramètres en apprentissage profond et apprentissage par transfert : applications en imagerie médicale." Electronic Thesis or Diss., Université Paris-Saclay (ComUE), 2019. http://www.theses.fr/2019SACLT001.
Full textIn the last few years, deep learning has changed irrevocably the field of computer vision. Faster, giving better results, and requiring a lower degree of expertise to use than traditional computer vision methods, deep learning has become ubiquitous in every imaging application. This includes medical imaging applications. At the beginning of this thesis, there was still a strong lack of tools and understanding of how to build efficient neural networks for specific tasks. Thus this thesis first focused on the topic of hyper-parameter optimization for deep neural networks, i.e. methods for automatically finding efficient neural networks on specific tasks. The thesis includes a comparison of different methods, a performance improvement of one of these methods, Bayesian optimization, and the proposal of a new method of hyper-parameter optimization by combining two existing methods: Bayesian optimization and Hyperband.From there, we used these methods for medical imaging applications such as the classification of field-of-view in MRI, and the segmentation of the kidney in 3D ultrasound images across two populations of patients. This last task required the development of a new transfer learning method based on the modification of the source network by adding new geometric and intensity transformation layers.Finally this thesis loops back to older computer vision methods, and we propose a new segmentation algorithm combining template deformation and deep learning. We show how to use a neural network to predict global and local transformations without requiring the ground-truth of these transformations. The method is validated on the task of kidney segmentation in 3D US images
Boyer, Laurent. "Apprentissage probabiliste de similarités d'édition." Phd thesis, Université Jean Monnet - Saint-Etienne, 2011. http://tel.archives-ouvertes.fr/tel-00718835.
Full textSoldano, Henry. "Apprentissage : Paradigmes, Structures et abstractions." Habilitation à diriger des recherches, Université Paris-Nord - Paris XIII, 2009. http://tel.archives-ouvertes.fr/tel-00514160.
Full textBondu, Alexis. "Apprentissage actif par modèles locaux." Phd thesis, Université d'Angers, 2008. http://tel.archives-ouvertes.fr/tel-00450124.
Full textTommasi, 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.
Full textÀ 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.
Decaestecker, Christine. "Apprentissage en classification conceptuelle incrémentale." Doctoral thesis, Universite Libre de Bruxelles, 1991. http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/213000.
Full textPellerin, Éric. "Méta-apprentissage des algorithmes génétiques /." Trois-Rivières : Université du Québec à Trois-Rivières, 2005. http://www.uqtr.ca/biblio/notice/resume/24652684R.pdf.
Full textLopez, Matthieu. "Apprentissage de problèmes de contraintes." Phd thesis, Université d'Orléans, 2011. http://tel.archives-ouvertes.fr/tel-00668156.
Full textPIERRE, CORINNE. "Apprentissage par l'action. Perspective developpementale." Paris 5, 1995. http://www.theses.fr/1995PA05H044.
Full textDe, Carvalho Gomes Fernando. "Utilisation d'algorithmes stochastiques en apprentissage." Montpellier 2, 1992. http://www.theses.fr/1992MON20254.
Full textRonald, Edmund. "Apprentissage évolutionniste des réseaux neuromimétiques." Palaiseau, Ecole polytechnique, 1997. http://www.theses.fr/1997EPXX0048.
Full textAl, Sahyouni Bou Fadel Reine. "TIC et apprentissage de l'interculturalité." Thesis, Bordeaux 3, 2014. http://www.theses.fr/2014BOR30020/document.
Full textThis thesis has aimed at conducting experimentally informed reflection on the role or the function of ICT in the process of acquisition of LE and learning of multiculturalism and openness to the other. It also aims to evaluate the potential of students’ acquisitionnel in knowledge, skills, communication and openness to other cultures. The research supported by this problem is also related to the role of ICT in the learning process in primary school: To what extent these new technologies involved in developing young people's skills in the field of French language and intercultural? ICT allow for simulations of various phenomena (physical, historical ...), they provide students with very diverse ways to realize the skills and knowledge they can develop while awakening their curiosity and motivation. Field research is based on two concrete learning experience language and French culture through networking school in various Francophone countries each using ICT. The method of work based on the observation of the practices of students and their evolution in the knowledge of the other where interculturalism , an analysis of web site content , surveys by semi -structured interviews and questionnaire with students and teachers
Jorro, José. "Identité, apprentissage et auto-organisation." Lille 3 : ANRT, 1986. http://catalogue.bnf.fr/ark:/12148/cb37594832v.
Full textGoh, Hanlin. "Apprentissage de Représentations Visuelles Profondes." Phd thesis, Université Pierre et Marie Curie - Paris VI, 2013. http://tel.archives-ouvertes.fr/tel-00948376.
Full textPellerin, Éric. "Méta-apprentissage des algorithmes génétiques." Thèse, Université du Québec à Trois-Rivières, 2005. http://depot-e.uqtr.ca/1805/1/000131549.pdf.
Full textBigot, Damien. "Représentation et apprentissage de préférences." Thesis, Toulouse 3, 2015. http://www.theses.fr/2015TOU30031/document.
Full text--
Tokmakov, Pavel. "Apprentissage à partir du mouvement." Thesis, Université Grenoble Alpes (ComUE), 2018. http://www.theses.fr/2018GREAM031/document.
Full textWeakly-supervised learning studies the problem of minimizing the amount of human effort required for training state-of-the-art models. This allows to leverage a large amount of data. However, in practice weakly-supervised methods perform significantly worse than their fully-supervised counterparts. This is also the case in deep learning, where the top-performing computer vision approaches remain fully-supervised, which limits their usage in real world applications. This thesis attempts to bridge the gap between weakly-supervised and fully-supervised methods by utilizing motion information. It also studies the problem of moving object segmentation itself, proposing one of the first learning-based methods for this task.We focus on the problem of weakly-supervised semantic segmentation. This is especially challenging due to the need to precisely capture object boundaries and avoid local optima, as for example segmenting the most discriminative parts. In contrast to most of the state-of-the-art approaches, which rely on static images, we leverage video data with object motion as a strong cue. In particular, our method uses a state-of-the-art video segmentation approach to segment moving objects in videos. The approximate object masks produced by this method are then fused with the semantic segmentation model learned in an EM-like framework to infer pixel-level semantic labels for video frames. Thus, as learning progresses, the quality of the labels improves automatically. We then integrate this architecture with our learning-based approach for video segmentation to obtain a fully trainable framework for weakly-supervised learning from videos.In the second part of the thesis we study unsupervised video segmentation, the task of segmenting all the objects in a video that move independently from the camera. This task presents challenges such as strong camera motion, inaccuracies in optical flow estimation and motion discontinuity. We address the camera motion problem by proposing a learning-based method for motion segmentation: a convolutional neural network that takes optical flow as input and is trained to segment objects that move independently from the camera. It is then extended with an appearance stream and a visual memory module to improve temporal continuity. The appearance stream capitalizes on the semantic information which is complementary to the motion information. The visual memory module is the key component of our approach: it combines the outputs of the motion and appearance streams and aggregates a spatio-temporal representation of the moving objects. The final segmentation is then produced based on this aggregated representation. The resulting approach obtains state-of-the-art performance on several benchmark datasets, outperforming the concurrent deep learning and heuristic-based methods
Barbot, Marie-José. "L'auto-apprentissage en milieu institutionnel." Paris 3, 1993. http://www.theses.fr/1994PA030069.
Full textThe system of auto-learning spreads out because of the double influence of evolution of behaviours and the technological possibilities. The discoveries in cognitive psychology and neurobiology emphasize the importance of the achievment of autonomy in learning. A survey among twenty-two auto-learners allowed to proove that the learners need the support that the etablisment should and must provide to them through a direct exposure that is to say the tutoring or mediated learning experience. Self-learning in institutional surrounding therefore does not mean the withdrawal of the teachers. On the contrary they must play a new part. That is why they must be appropriately trained to this aim
Béthune, Louis. "Apprentissage profond avec contraintes Lipschitz." Electronic Thesis or Diss., Université de Toulouse (2023-....), 2024. http://www.theses.fr/2024TLSES014.
Full textThis thesis explores the characteristics and applications of Lipschitz networks in machine learning tasks. First, the framework of "optimization as a layer" is presented, showcasing various applications, including the parametrization of Lipschitz-constrained layers. Then, the expressiveness of these networks in classification tasks is investigated, revealing an accuracy/robustness tradeoff controlled by entropic regularization of the loss, accompanied by generalization guarantees. Subsequently, the research delves into the utilization of signed distance functions as a solution to a regularized optimal transport problem, showcasing their efficacy in robust one-class learning and the construction of neural implicit surfaces. After, the thesis demonstrates the adaptability of the back-propagation algorithm to propagate bounds instead of vectors, enabling differentially private training of Lipschitz networks without incurring runtime and memory overhead. Finally, it goes beyond Lipschitz constraints and explores the use of convexity constraint for multivariate quantiles
Léon, Aurélia. "Apprentissage séquentiel budgétisé pour la classification extrême et la découverte de hiérarchie en apprentissage par renforcement." Electronic Thesis or Diss., Sorbonne université, 2019. http://www.theses.fr/2019SORUS226.
Full textThis thesis deals with the notion of budget to study problems of complexity (it can be computational complexity, a complex task for an agent, or complexity due to a small amount of data). Indeed, the main goal of current techniques in machine learning is usually to obtain the best accuracy, without worrying about the cost of the task. The concept of budget makes it possible to take into account this parameter while maintaining good performances. We first focus on classification problems with a large number of classes: the complexity in those algorithms can be reduced thanks to the use of decision trees (here learned through budgeted reinforcement learning techniques) or the association of each class with a (binary) code. We then deal with reinforcement learning problems and the discovery of a hierarchy that breaks down a (complex) task into simpler tasks to facilitate learning and generalization. Here, this discovery is done by reducing the cognitive effort of the agent (considered in this work as equivalent to the use of an additional observation). Finally, we address problems of understanding and generating instructions in natural language, where data are available in small quantities: we test for this purpose the simultaneous use of an agent that understands and of an agent that generates the instructions
Kozlova, Olga. "Apprentissage par renforcement hiérarchique et factorisé." Phd thesis, Université Pierre et Marie Curie - Paris VI, 2010. http://tel.archives-ouvertes.fr/tel-00632968.
Full textQuerrec, Ronan. "Apprentissage de procédures en environnements virtuels." Habilitation à diriger des recherches, Université Européenne de Bretagne, 2010. http://tel.archives-ouvertes.fr/tel-00557039.
Full textFilippi, Sarah. "Stratégies optimistes en apprentissage par renforcement." Phd thesis, Ecole nationale supérieure des telecommunications - ENST, 2010. http://tel.archives-ouvertes.fr/tel-00551401.
Full textCleeremans, Axel. "Conscience et apprentissage: une perspective dynamique." Doctoral thesis, Universite Libre de Bruxelles, 2001. http://hdl.handle.net/2013/ULB-DIPOT:oai:dipot.ulb.ac.be:2013/211510.
Full textMaillard, Odalric-Ambrym. "APPRENTISSAGE SÉQUENTIEL : Bandits, Statistique et Renforcement." Phd thesis, Université des Sciences et Technologie de Lille - Lille I, 2011. http://tel.archives-ouvertes.fr/tel-00845410.
Full textAugier, Sébastien. "Apprentissage Supervisé Relationnel par Algorithmes d'Évolution." Phd thesis, Université Paris Sud - Paris XI, 2000. http://tel.archives-ouvertes.fr/tel-00947322.
Full textGandar, Benoît. "Apprentissage actif pour l'approximation de variétés." Phd thesis, Université Blaise Pascal - Clermont-Ferrand II, 2012. http://tel.archives-ouvertes.fr/tel-00954409.
Full textMoura, Simon. "Apprentissage multi-cibles : théorie et applications." Thesis, Université Grenoble Alpes (ComUE), 2018. http://www.theses.fr/2018GREAM085.
Full textIn this thesis, we study the problem of learning with multiple outputs related to different tasks, such as classification and ranking. In this line of research, we explored three different axes. First we proposed a theoretical framework that can be used to show the consistency of multi-label learning in the case of classifier chains, where outputs are homogeneous. Based on this framework, we proposed Rademacher generalization error bound made by any classifier in the chain and exhibit dependency factors relating each output to the others. As a result, we introduced multiple strategies to learn classifier chains and select an order for the chain. Still focusing on the homogeneous multi-output framework, we proposed a neural network based solution for fine-grained sentiment analysis and show the efficiency of the approach. Finally, we proposed a framework and an empirical study showing the interest of learning with multiple tasks, even when the outputs are of different types
Martinez, Medina Lourdes. "Optimisation des requêtes distribuées par apprentissage." Thesis, Grenoble, 2014. http://www.theses.fr/2014GRENM015.
Full textDistributed data systems are becoming increasingly complex. They interconnect devices (e.g. smartphones, tablets, etc.) that are heterogeneous, autonomous, either static or mobile, and with physical limitations. Such devices run applications (e.g. virtual games, social networks, etc.) for the online interaction of users producing / consuming data on demand or continuously. The characteristics of these systems add new dimensions to the query optimization problem, such as multi-optimization criteria, scarce information on data, lack of global system view, among others. Traditional query optimization techniques focus on semi (or not at all) autonomous systems. They rely on information about data and make strong assumptions about the system behavior. Moreover, most of these techniques are centered on the optimization of execution time only. The difficulty for evaluating queries efficiently on nowadays applications motivates this work to revisit traditional query optimization techniques. This thesis faces these challenges by adapting the Case Based Reasoning (CBR) paradigm to query processing, providing a way to optimize queries when there is no prior knowledge of data. It focuses on optimizing queries using cases generated from the evaluation of similar past queries. A query case comprises: (i) the query, (ii) the query plan and (iii) the measures (computational resources consumed) of the query plan. The thesis also concerns the way the CBR process interacts with the query plan generation process. This process uses classical heuristics and makes decisions randomly (e.g. when there are no statistics for join ordering and selection of algorithms, routing protocols). It also (re)uses cases (existing query plans) for similar queries parts, improving the query optimization, and therefore evaluation efficiency. The propositions of this thesis have been validated within the CoBRa optimizer developed in the context of the UBIQUEST project