Tesis sobre el tema "Apprentissage de systèmes dynamiques"
Crea una cita precisa en los estilos APA, MLA, Chicago, Harvard y otros
Consulte los 50 mejores tesis para su investigación sobre el tema "Apprentissage de systèmes dynamiques".
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.
Ralaivola, Liva. "Modélisation et apprentissage de systèmes et de concepts dynamiques". Paris 6, 2003. http://www.theses.fr/2003PA066277.
Texto completoJEANPIERRE, Laurent. "Apprentissage et adaptation pour la modélisation stochastique de systèmes dynamiques réels". Phd thesis, Université Henri Poincaré - Nancy I, 2002. http://tel.archives-ouvertes.fr/tel-00003378.
Texto completoJeanpierre, Laurent. "Apprentissage et adaptation pour la modélisation stochastique de systèmes dynamiques réels". Nancy 1, 2002. http://docnum.univ-lorraine.fr/public/SCD_T_2002_0246_JEANPIERRE.pdf.
Texto completoThe exploitation of Artificial Intelligence algorithms in real conditions is an interesting method for their improvement, since weaknesses are shown very quickly thanks to real, uncontrolled constraints. In particular, I study two problems of medical diagnosis and a classical problem of robot navigation. Using fuzzy sets with Markov models provide an intuitive but powerful system to solve such situations. Then, I introduce diagnosis learning which betters the cooperation with doctors, as it allows correcting the model while ensuring numerical stability. Thus, doctors can modify the patient model without setting each parameter manually. Finally, I show this approach can be generalized to a whole class of diagnosis problems. To achieve this goal, I show an integrated development environment that allows to simply link modules altogether to have a given problem solved. This should help creating new applications, while minimizing the programming time of researchers
Franceschi, Jean-Yves. "Apprentissage de représentations et modèles génératifs profonds dans les systèmes dynamiques". Electronic Thesis or Diss., Sorbonne université, 2022. http://www.theses.fr/2022SORUS014.
Texto completoThe recent rise of deep learning has been motivated by numerous scientific breakthroughs, particularly regarding representation learning and generative modeling. However, most of these achievements have been obtained on image or text data, whose evolution through time remains challenging for existing methods. Given their importance for autonomous systems to adapt in a constantly evolving environment, these challenges have been actively investigated in a growing body of work. In this thesis, we follow this line of work and study several aspects of temporality and dynamical systems in deep unsupervised representation learning and generative modeling. Firstly, we present a general-purpose deep unsupervised representation learning method for time series tackling scalability and adaptivity issues arising in practical applications. We then further study in a second part representation learning for sequences by focusing on structured and stochastic spatiotemporal data: videos and physical phenomena. We show in this context that performant temporal generative prediction models help to uncover meaningful and disentangled representations, and conversely. We highlight to this end the crucial role of differential equations in the modeling and embedding of these natural sequences within sequential generative models. Finally, we more broadly analyze in a third part a popular class of generative models, generative adversarial networks, under the scope of dynamical systems. We study the evolution of the involved neural networks with respect to their training time by describing it with a differential equation, allowing us to gain a novel understanding of this generative model
Veillon, Lise-Marie. "Apprentissage artificiel collectif ; aspects dynamiques et structurels". Thesis, Sorbonne Paris Cité, 2018. http://www.theses.fr/2018USPCD004/document.
Texto completoCollective learning in multi-agent systems considers how a community of autonomous agents sharing a learning purpose may benefit from exchanging information to learn efficiently as a community as well as individuals. The community forms a communication network where each agent may accesses observations, called learning examples. This thesis is based on a former protocol, SMILE (Sound-Multi-agent-Incremental-LEarning), which sets up parsimonious examples and hypotheses exchanges between agents. In a fully connected community, this protocol guarantees an agent’s hypothesis takes into account all the examples obtained by the community. Some sequential protocols add propagation to SMILE in order to extend this consistency guarantee to other connected networks. This thesis contribution to the artificial collective learning field is two fold.First, we investigate the influence of network structures on learning in networks when communication is limited to neighbourhood without further information propagation. Second, we present and analyze a new protocol, Waves, with SMILE’s guarantees and a more dynamic learning process thanks to its execution in parallel. The evaluation of this protocol in a simple turn-based setting gives the opportunity to improve it here in multiple ways. It is however meant to be used with online learning without any restriction on the acquisition rate of new examples, neither on speed nor number
Massé, Pierre-Yves. "Autour De L'Usage des gradients en apprentissage statistique". Thesis, Université Paris-Saclay (ComUE), 2017. http://www.theses.fr/2017SACLS568/document.
Texto completoWe prove a local convergence theorem for the classical dynamical system optimization algorithm called RTRL, in a nonlinear setting. The rtrl works on line, but maintains a huge amount of information, which makes it unfit to train even moderately big learning models. The NBT algorithm turns it by replacing these informations by a non-biased, low dimension, random approximation. We also prove the convergence with arbitrarily close to one probability, of this algorithm to the local optimum reached by the RTRL algorithm. We also formalize the LLR algorithm and conduct experiments on it, on synthetic data. This algorithm updates in an adaptive fashion the step size of a gradient descent, by conducting a gradient descent on this very step size. It therefore partially solves the issue of the numerical choice of a step size in a gradient descent. This choice influences strongly the descent and must otherwise be hand-picked by the user, following a potentially long research
Canu, Michaël. "Apport de l'étude conjointe de systèmes dynamiques libres et commandés dans la compréhension des concepts d'équilibre et de stabilité". Paris 7, 2014. http://www.theses.fr/2014PA070099.
Texto completoThe concepts of balance and stability are very important in control and more generally in ail the domains usin a systematic approach aiming, among others, at the understanding or at the implementation of mechanisms of regulation (biology, economy, electronics, chemistry, etc. ). We noticed a relative lack of understanding of these concepts at the students, including after control systems courses, either at university and engineering school. The research in science education was interested a lot in the understanding of the concept of equilibrium in chemistry, but enough Utile in mechanics for example, and essentially from an academic declarative and procedural knowledge point of view. Our work shows that the conceptions of the students relative to this concept call on, in a substantial way, to the non-academic domains and especially to the concept of stability. On one hand, our work consisted in trying to understand the reasons of the difficulties observed from the study of the students' conceptions (and possible factors which could influence their development) and on the other hand, to propose a classroom sequence at the engineer's education level, to promote a conceptual change, by using the methodological frame of the didactic engineering (Artigue) and the theoretical frame of the conceptual camps. We show that it is possible to obtain an improvement of the understanding of these concepts with this kind of teaching sequence
Ramdani, Mohammed. "Système d'induction formelle à base de connaissances imprécises". Paris 6, 1994. http://www.theses.fr/1994PA066237.
Texto completoTeulier, Caroline. "Nature des transitions dans l'évolution des coordinations lors de l'apprentissage d'habiletés complexes". Montpellier 1, 2005. http://www.theses.fr/2005MON14005.
Texto completoDaucé, Emmanuel. "Adaptation dynamique et apprentissage dans les réseaux de neurones récurrents aléatoires". Toulouse, ENSAE, 2000. https://tel.archives-ouvertes.fr/tel-01394004.
Texto completoNachouki, Mirna. "L'acquisition de connaissances dans les systèmes dynamiques : production et utilisation dans le cadre de l'atelier de génie didacticiel intégré". Toulouse 3, 1995. http://www.theses.fr/1995TOU30001.
Texto completoPlassart, Stéphan. "Optimisation en-ligne pour les systèmes dynamiques en temps-réel". Thesis, Université Grenoble Alpes, 2020. http://www.theses.fr/2020GRALM017.
Texto completoThe energy consumption is a crucial issue for real-time systems,that's why optimizing it online, i.e. while the processor is running, has become essential and will be the goal of this thesis.This optimization is done by adapting the processor speed during the job execution.This thesis addresses several situations with different knowledge on past, active and future job characteristics.Firstly, we consider that all job characteristics are known (the offline case),and we propose a linear time algorithm to determine the speed schedule to execute n jobs on a single processor.Secondly, using Markov decision processes, we solve the case where past and active job characteristics are entirely known,and for future jobs only the probability distribution of the jobs characteristics (arrival times, execution times and deadlines) are known.Thirdly we study a more general case: the execution is only discovered when the job is completed.In addition we also consider the case where we have no statistical knowledge on jobs,so we have to use learning methods to determine the optimal processor speeds online.Finally, we propose a feasibility analysis (the processor ability to execute all jobs before its deadline when it works always at maximal speed) of several classical online policies,and we show that our dynamic programming algorithm is also the best in terms of feasibility
Scesa, Vincent. "Contrôleurs neuronaux dynamiques et apprentissage : du perfectionnement des algorithmes à leur application temps réel sur des systèmes robotiques". Versailles-St Quentin en Yvelines, 2006. http://www.theses.fr/2006VERS0009.
Texto completoHow can we control a system that is difficult to model or with an incomplete knowledge?When theorical modeling reaches its limits, the cooperation between observation, learning and experience could answer this question and fill the lack of conventional techniques. This thesis gives a complete view of how to carry out this kind of control methodology for complex dynamical systems. Through a proposed classification of the current control techniques, the algorithm choices are first exposed (CTRNN networks - Continuous Time Recurrent Neural Networks - and BPTT learning -BackPropagation Through Time). To enable the stabilization and the amelioration of their behaviour, adaptations and improvements are then proposed and experimentally validated. Next, the algorithmic principles carried out to follow the real time constraints of our applications shows a way to implement this controller in industrial electronic units. At last, the experiments carried out on two polyarticulated dynamical systems: the road simulator of the BIA company and the biped robot of the LISV laboratory, confronts the developed methodology with real applications and associated constraints. A discussion on the use of our controller for the command of this kind of system is finally given
Boukharouba, Khaled. "Modélisation et classification de comportements dynamiques des systèmes hybrides". Thesis, Lille 1, 2011. http://www.theses.fr/2011LIL10088/document.
Texto completoIn this thesis, we consider the identification of a special class of hybrid systems which is the class of PieceWise Affine (PWA) systems from input-output data. The identification of PWA models is a challenging problem. It involves the estimation of both the parameters of the affine sub-models, and the coefficients of the hyperplanes defining the partition of the state + input set. First, we give an overview of the different approaches available in the literature for the identification of PWA systems. Then, we propose new methods for identifying PWA models from data. The solution includes the estimation of the number of sub-models, the identification of the parameter vectors that describe the different sub-models and the determination of the bounding hyperplanes of the polyhedral regions associated with the sub-models. After this, we propose a recursive algorithm for identifying PieceWise Affine systems (PWA) and PieceWise nonlinear systems where the parameters of the sub-models and the regions can vary over time. A recursive LS-SVM technique is proposed for recursive updating of the parameters of each sub-model. The adaptation of the parameters of the regions is ensured by an online multi-category support vector classifier. The last part of this work is devoted to the validation of our methods on real examples. We apply our methods to the identification of a hydraulic system and a pick-and-place machine. We also show how the temporal segmentation of video into different shots can be performed, based on the estimation of local linear sub-models
Mezine, Adel. "Conduite d'expériences par apprentissage actif pour l'identification de systèmes dynamiques biologiques : application à l'estimation de paramètres d'équations différentielles ordinaires". Thesis, Université Paris-Saclay (ComUE), 2016. http://www.theses.fr/2016SACLE030/document.
Texto completoContinuous progress in screening and high-throughput sequencing techniques in recent years paves the way for the identification of dynamic biological systems such as gene regulatory networks. However, the scarcity of the experimental data often leads to anuncertain estimation of parameters of interest. These uncertainties can be solved by generating new data in different experimental conditions, which induces additional costs. This thesis proposes a general active learning approach to develop tools of sequential experimental design for the identification of dynamical biological systems. The problem is formulated as a one-player game : the player has a budget dedicated for his experiments, each experiment has a different cost ; at every turn, he chooses one or more experiments to be performed on the system with the ultimate aim of maximizing the quality of the estimate, until the available budget is exhausted. The proposed approach called Experimental DEsign for Network inference (EDEN), is based on UCT (Upper Confident bounds for Trees) algorithm which combines Monte-Carlo tree search algorithms with multi-arm bandits to perform an effective exploration of the possible sequences of experiments. A strong point of the approach is anticipation : an experiment is selected at each round, knowing that this round will be followed by a number of other experiments, according to the available budget. This generic approach is rolled out in parameter estimation in nonlinear ordinary differential equations using partial observations. EDEN is applied on two problems : signaling network and gene regulatory network identification. Compared to the competitors, it exhibits very good results on a DREAM7 challenge where a limited budget and a wide range of experiments (perturbations, measurements) are available
Dzogang, Fabon. "Représentation et apprentissage à partir de textes pour des informations émotionnelles et pour des informations dynamiques". Paris 6, 2013. http://www.theses.fr/2013PA066253.
Texto completoAutomatic knowledge extraction from texts consists in mapping lowlevel information, as carried by the words and phrases extracted fromdocuments, to higher level information. The choice of datarepresentation for describing documents is, thus, essential and thedefinition of a learning algorithm is subject to theirspecifics. This thesis addresses these two issues in the context ofemotional information on the one hand and dynamic information on theother. In the first part, we consider the task of emotion extraction forwhich the semantic gap is wider than it is with more traditionalthematic information. Therefore, we propose to study representationsaimed at modeling the many nuances of natural language used fordescribing emotional, hence subjective, information. Furthermore, wepropose to study the integration of semantic knowledge which provides,from a characterization perspective, support for extracting theemotional content of documents and, from a prediction perspective,assistance to the learning algorithm. In the second part, we study information dynamics: any corpus ofdocuments published over the Internet can be associated to sources inperpetual activity which exchange information in a continuousmovement. We explore three main lines of work: automaticallyidentified sources; the communities they form in a dynamic and verysparse description space; and the noteworthy themes they develop. Foreach we propose original extraction methods which we apply to a corpusof real data we have collected from information streams over the Internet
Colliaux, David. "Classes of neuronal dynamics and experience dependent structured correlations in the visual cortex". Palaiseau, Ecole polytechnique, 2011. http://pastel.archives-ouvertes.fr/docs/00/67/61/04/PDF/HPEA.pdf.
Texto completoL'activité neuronale est souvent considérée en neuroscience cognitive par la réponse évoquée mais l'essentiel de l'énergie consommée par le cerveau permet d'entretenir les dynamiques spontanées des réseaux corticaux. L'utilisation combinée d'algorithmes de classification (K means, arbre hirarchique, SOM) sur des enregistrements intracellulaires du cortex visuel primaire du chat nous permet de définir des classes de dynamiques neuronales et de les comparer l'activité évoquée par un stimulus visuel. Ces dynamiques peuvent être étudiées sur des systèmes simplifiés (FitzHugh-Nagumo, systèmes dynamiques hybrides, Wilson-Cowan) dont nous présentons l'analyse. Enfin, par des simulations de réseaux composés de colonnes de neurones, un modèle du cortex visuel primaire nous permet d'étudier les dynamiques spontanées et leur effet sur la réponse à un stimulus. Après une période d'apprentissage pendant laquelle des stimuli visuels sont presentés, des vagues de dépolarisation se propagent dans le réseau. L'étude des correlations dans ce réseau montre que les dynamiques spontanées reflètent les propriétés fonctionnelles acquises au cours de l'apprentissage
Komar, John. "Dynamique de l'apprentissage moteur : apprendre loin de l'équilibre". Rouen, 2013. http://www.theses.fr/2013ROUEL009.
Texto completoBased on the theoretical background of Dynamical Systems Theory, the aim of this thesis is to analyze the temporal dynamics of motor coordination, in relation to the constraints acting on the learner. Specifically, this work investigates the learning process of arm-leg coordination in aquatic locomotion, notably in breaststroke swimming, which is a context where natural individual motor behavior is specifically constrained by environmental constraints. An emphasis is placed on the study of the functional rôle of coordination variability and the rôle of manipulating constraints as a way of promoting exploratory activity in order to investigate issues surrounding learning optimization. Through the use of 3-Dimensions video analysis and motion sensors, the arm-leg coordination has been studied based on the oscillations of knee and elbow angles. The stability and the flexibility of the expert pattern of coordination has been highlighted (study 1), as well as the effect of the level of environmental constraints on the expression of neurobiological degeneracy (study 2). The use of verbal instructions as temporary constraints acting on the learner has then been studied through his effect on the temporal dynamics of the coordination and the different search strategies exhibited by learners (studies 3 and 4). Overall, The results tend to show that in a specific environment as the aquatic environment, the to-be-learned pattern of coordination refers to a biomechanically expert coordination. Nevertheless, the qualitative changes during learning operate through periods of metastability, reflecting an active exploration of the perceptual-motor workspace. In addition, manipulating task constraints without prescribing the to-be-learned pattern can be useful in guiding the exploration (i. E. Proscribe what not to do rather than prescribe what to do). These results advocate for the use of a non-linear pedagogical approach that tends to encourage exploratory learning
CERRADA, LOZADA Mariela. "Sur les modèles flous adaptatifs dynamiques". Phd thesis, INSA de Toulouse, 2003. http://tel.archives-ouvertes.fr/tel-00010013.
Texto completoBézenac, Emmanuel de. "Modeling physical processes with deep learning : a dynamical systems approach". Electronic Thesis or Diss., Sorbonne université, 2021. http://www.theses.fr/2021SORUS203.
Texto completoDeep Learning has emerged as a predominant tool for AI, and has already abundant applications in fields where data is abundant and access to prior knowledge is difficult. This is not necessarily the case for natural sciences, and in particular, for physical processes. Indeed, these have been the object of study since centuries, a vast amount of knowledge has been acquired, and elaborate algorithms and methods have been developped. Thus, this thesis has two main objectives. The first considers the study of the role that deep learning has to play in this vast ecosystem of knowledge, theory and tools. We will attempt to answer this general question through a concrete problem: the one of modelling complex physical processes, leveraging deep learning methods in order to make up for lacking prior knowledge. The second objective is somewhat its converse: it focuses on how perspectives, insights and tools from the field of study of physical processes and dynamical systems can be applied in the context of deep learning, in order to gain a better understanding and develop novel algorithms
Ben, Abdallah Emna. "Étude de la dynamique des réseaux biologiques : apprentissage des modèles, intégration des données temporelles et analyse formelle des propriétés dynamiques". Thesis, Ecole centrale de Nantes, 2017. http://www.theses.fr/2017ECDN0041.
Texto completoOver the last few decades, the emergence of a wide range of new technologies has produced a massive amount of biological data (genomics, proteomics...). Thus, a very large amount of time series data is now produced every day. The newly produced data can give us new ideas about the behavior of biological systems. This leads to considerable developments in the field of bioinformatics that could benefit from these enormous data. This justifies the motivation to develop efficient methods for learning Biological Regulatory Networks (BRN) modeling a biological system from its time series data. Then, in order to understand the nature of system functions, we study, in this thesis, the dynamics of their BRN models. Indeed, we focus on developing original and scalable logical methods (implemented in Answer Set Programming) to deciphering the emerging complexity of dynamics of biological systems. The main contributions of this thesis are enumerated in the following. (i) Refining the dynamics of the BRN, modeling with the automata Network (AN) formalism, by integrating a temporal parameter (delay) in the local transitions of the automata. We call the extended formalism a Timed Automata Network (T-AN). This integration allows the parametrization of the transitions between each automata local states as well as between the network global states. (ii) Learning BRNs modeling biological systems from their time series data. (iii) Model checking of discrete dynamical properties of BRN (modeling with AN and T-AN) by dynamical formal analysis : attractors identification (minimal trap domains from which the network cannot escape) and reachability verification of an objective from a network global initial state
Lepère, Stéphane. "Contribution à la prédiction en ligne des séries temporelles : un cas d'étude à la modélisation de systèmes dynamiques". Lille 1, 2001. https://pepite-depot.univ-lille.fr/RESTREINT/Th_Num/2001/50376-2001-219.pdf.
Texto completoDia, Diyé. "Systèmes producteurs de confiance : ouverture de droit à des services par apprentissage dynamique du comportement des utilisateurs du système d'information". Thesis, Clermont-Ferrand 2, 2016. http://www.theses.fr/2016CLF22396/document.
Texto completoRolland, de Rengervé Antoine. "Apprentissage Interactif en Robotique Autonome : vers de nouveaux types d'IHM". Phd thesis, Université de Cergy Pontoise, 2013. http://tel.archives-ouvertes.fr/tel-00969519.
Texto completoLe, Van Luong. "Identification de systèmes dynamiques hybrides : géométrie, parcimonie et non-linéarités". Phd thesis, Université de Lorraine, 2013. http://tel.archives-ouvertes.fr/tel-00874283.
Texto completoDonà, Jérémie. "Statistical learning of physical dynamics". Electronic Thesis or Diss., Sorbonne université, 2022. http://www.theses.fr/2022SORUS166.
Texto completoThe modeling of natural processes relies on a physical description that prescribes the changes in the state of the studied system. The use of domain specific knowledge about the system allows the translation of physical principles into models, which are then validated by experimental data. With its successes in many domain like image classification, deep learning has become a powerful tool for the modeling of physical processes, thanks to the significant increase in the amount of data available from sensors. Statistical learning of physical processes by a sole data-driven approach suffers from several limitations such as interpretation difficulties, stability during training and reduced generalization capabilities. The objective of this work is to provide tools in order to perform data-driven learning of physical processes. In particular, we study spatio-temporal phenomena which dynamics obey a differential equation and focus on incorporating domain and physical knowledge in learning algorithms. This leads us to study hybrid physical-statistical systems for the modeling of physical processes. We will identify the problems related to the learning of hybrid dynamics and propose a framework including constraints adapted to deep networks to improve the interpretability and the performance of the learned algorithms. Conversely, dynamical systems have provided numerous tools to improve statistical models. However, neural networks remain qualified as "black boxes" because they are not interpretable. Thus, we will attempt to open the black box and propose more interpretable neural network architectures with increased generalization performances for the modeling of spatio-temporal systems
Rolland, de Rengerve Antoine. "Apprentissage Interactif en Robotique Autonome : vers de nouveaux types d'IHM". Thesis, Cergy-Pontoise, 2013. http://www.theses.fr/2013CERG0664/document.
Texto completoAn autonomous robot collaborating with humans should be able to learn how to navigate and manipulate objects in the same task. In a classical approach, independent functional modules are considered to manage the different aspects of the task (navigation, arm control,...) . To the contrary, the goal of this thesis is to show that learning tasks of different kinds can be tackled by learning sensorimotor attractors from a few task nonspecific structures. We thus proposed an architecture which can learn and encode attractors to perform navigation tasks as well as arm control.We started by considering a model inspired from place-cells for navigation of autonomous robots. On-line and interactive learning of place-action couples can let attraction basins emerge, allowing an autonomous robot to follow a trajectory. The robot behavior can be corrected and guided by interacting with it. The successive corrections and their sensorimotor coding enables to define the attraction basin of the trajectory. My first contribution was to adapt this principle of sensorimotor attractor building for the impedance control of a robot arm. While a proprioceptive posture is maintained, the arm movements can be corrected by modifying on-line the motor command expressed as muscular activations. The resulting motor attractors are simple associations between the proprioceptive information of the arm and these motor commands. I then showed that the robot could learn visuomotor attractors by combining the proprioceptive and visual information with the motor attractors. The visuomotor control corresponds to a homeostatic system trying to maintain an equilibrium between the two kinds of information. In the case of ambiguous visual information, the robot may perceive an external stimulus (e.g. a human hand) as its own hand. According to the principle of homeostasis, the robot will act to reduce the incoherence between this external information and its proprioceptive information. It then displays a behavior of immediately observed gestures imitation. This mechanism of homeostasis, completed by a memory of the observed sequences and action inhibition capability during the observation phase, enables a robot to perform deferred imitation and learn by observation. In the case of more complex tasks, we also showed that learning transitions can be the basis for learning sequences of gestures, like in the case of cognitive map learning in navigation. The use of motivational contexts then enables to choose between different learned sequences.We then addressed the issue of integrating in the same architecture behaviors involving visuomotor navigation and robotic arm control to grab objects. The difficulty is to be able to synchronize the different actions so the robot act coherently. Erroneous behaviors of the robot are detected by evaluating the actions predicted by the model with respect to corrections forced by the human teacher. These situations can be learned as multimodal contexts modulating the action selection process in order to adapt the behavior so the robot reproduces the desired task.Finally, we will present the perspectives of this work in terms of sensorimotor control, for both navigation and robotic arm control, and its link to human robot interface issues. We will also insist on the fact that different kinds of imitation behavior can result from the emergent properties of a sensorimotor control architecture
Quoy, Mathias. "Apprentissage dans les réseaux neuromimétiques à dynamique chaotique". Toulouse, ENSAE, 1994. http://www.theses.fr/1994ESAE0009.
Texto completoPierrot, David. "Détection dynamique des intrusions dans les systèmes informatiques". Thesis, Lyon, 2018. http://www.theses.fr/2018LYSE2077.
Texto completoThe expansion and democratization of the digital world coupled with the effect of the Internet globalization, has allowed individuals, countries, states and companies to interconnect and interact at incidence levels never previously imagined. Cybercrime, in turn, is unfortunately one the negative aspects of this rapid global interconnection expansion. We often find malicious individuals and/or groups aiming to undermine the integrity of Information Systems for either financial gain or to serve a cause. The consequences of an intrusion can be problematic for the existence of a company or an organization. The impacts are synonymous with financial loss, brand image degradation and lack of seriousness. The detection of an intrusion is not an end in itself, the reduction of the delta detection-reaction has become a priority. The different existing solutions prove to be cumbersome to set up. Research has identified more efficient data mining methods, but integration into an information system remains difficult. Capturing and converting protected resource data does not allow detection within acceptable time frames. Our contribution helps to detect intrusions. Protect us against Firewall events which reduces the need for computing power while limiting the knowledge of the information system by intrusion detectors. We propose an approach taking into account the technical aspects by the use of a hybrid method of data mining but also the functional aspects. The addition of these two aspects is grouped into four phases. The first phase is to visualize and identify network activities. The second phase concerns the detection of abnormal activities using data mining methods on the source of the flow but also on the targeted assets. The third and fourth phases use the results of a risk analysis and a safety verification technique to prioritize the actions to be carried out. All these points give a general vision on the hygiene of the information system but also a direction on monitoring and corrections to be made.The approach developed to a prototype named D113. This prototype, tested on a platform of experimentation in two architectures of different size made it possible to validate our orientations and approaches. The results obtained are positive but perfectible. Prospects have been defined in this direction
Guivarch, Valérian. "Prise en compte de la dynamique du contexte pour les systèmes ambiants par systèmes multi-agents adaptatifs". Toulouse 3, 2014. http://thesesups.ups-tlse.fr/2461/.
Texto completoThe ambient systems are composed by many heteregeneous devices, distributed in the environment, and interacting dynamically. So, the person is a central concern of these systems that have to adapt themselves to the users' context. Thos kind of systems are called/named context aware system. However, the strong dynamic of ambient systems makes impossible to design a priori all adaptation rules needed. The learning of the behaviour to give to an ambient system depending of its context, independantly of any a priori knowledge -knowledge about the behaviour he has to learn, about the used data, or about the users preferences- is the challenge to which this thesis tries to answer. The main contribution of this work is the design of the adaptive multi agent system Amadeus. Its objective is to learn a pertinent behaviour for an ambient system based on the observation of the reccuring actions performed by users, and then to determine in which contexts theses actions are performed in order to perform them on behalf of the user. The learning performed by Amadeus is based on the AMAS approach (Adaptive Multi-Agent System), and is local to each device. It consists in distributing and integrating the Amadeus agents to each device of the ambient system, these agents being able to determine locally and cooperatively the good behaviour to assign to the associated device depending of the users actions
Mondou, Damien. "Gestion adaptative des contenus numériques : proposition d’un framework générique par apprentissage et re-scénarisation dynamique". Thesis, La Rochelle, 2019. http://www.theses.fr/2019LAROS029.
Texto completoThis thesis aims to propose an architecture that addresses the design, supervision, management and adaptation of an interactive experience. We therefore propose a complete framework to facilitate the modeling phase of an interactive system and guarantee sufficient flexibility to achieve the objectives of complexity, scalability, adaptability and improvement through automatic learning. For this purpose, the formal model, CIT, based on two layers of description was introduced. The dynamic supervision process consists in controlling the interactive experience with regard to the formal model, based on networks of timed input/output automata. Two softwares, CELTIC (Common Editor for Location Time Interaction and Content) and EDAIN (Execution Driver based on Artificial INtelligence), implementing the CIT model and the activity supervision engine respectively, were developed during this thesis
ZEMOURI, RYAD. "Contribution à la surveillance des systèmes de production à l'aide des réseaux de neurones dynamiques : Application à la e-maintenance". Phd thesis, Université de Franche-Comté, 2003. http://tel.archives-ouvertes.fr/tel-00006003.
Texto completoBouguelid, Mohamed Saïd. "Contribution à l’application de la reconnaissance des formes et la théorie des possibilités au diagnostic adaptatif et prédictif des systèmes dynamiques". Reims, 2007. http://theses.univ-reims.fr/exl-doc/GED00000741.pdf.
Texto completoThe problem of diagnosis by Pattern Recognition can be posed as a problem of classification, i. E. , the actual functioning mode can be determined by knowing the class of the actual pattern. We use the method Fuzzy Pattern Matching (FPM) to realize the diagnosis because it is a simple method based on a feature selection. In addition it has a small and constant classification time, and it takes into account both the imprecision and uncertainty. However FPM is marginal, i. E. , its global decision is based on the selection of one of the intermediate decisions. Each intermediate decision is based on one attribute. Thus, FPM does not take into account the correlation between attributes. Additionally, FPM considers the shape of classes as convex one. Also, FPM cannot realize the adaptive and predictive diagnosis because it rejects all the points which carry the information about the class evolution or the creation of a new class. These drawbacks make FPM unusable for many real world applications. In this thesis, we propose to improve FPM to solve these drawbacks. Several synthetic and real data sets are used to show the performances of the improved FPM with respect to classical one
Martinez, Regis. "Dynamique des systèmes cognitifs et des systèmes complexes : étude du rôle des délais de transmission de l’information". Thesis, Lyon 2, 2011. http://www.theses.fr/2011LYO20054/document.
Texto completoHow memory information is represented is still an open question in neurobiology, but also, from the computer science point of view, in machine learning. Some artificial neuron networks models have to face the problem of retrieving information, knowing that, in regard to the model performance, this information is actually stored but in an unknown form or too complex to be easily accessible. This is one of the problems met in large neuron networks and which « reservoir computing » intends to answer.« Reservoir computing » is a category of models that has emerged at the same period as, and has propoerties similar to the model we present here. It is composed of three parts that are (1) an input layer that allows to inject learning examples, (2) a « reservoir » composed of neurons connected with or without a particular predefined, and where there can be adaptation mecanisms, (3) an output layer, called « readout », on which a supervised learning if performed. We bring a particularity that consists in using axonal delays, the propagation time of information from one neuron to another through an axonal connexion. Using delays is a computational improvement in the light of machin learning but also a biological argument for information representation.We show that our model is capable of a improvable but efficient and promising artificial learning. Based on this observation and in the aim of improving performance we seek to understand the internal dynamics of the model. More precisely we study how the topology of the reservoir can influence the dynamics. To do so, we make use of the theory of polychronous groups. We have developped complexe algorithms allowing us to detect those topologicodynamic structures in a network, and in a network activity having a given topology.If we succeed in understanding the links between topology and dynamics, we may take advantage of it to be able to create reservoir with specific properties, suited for learning. Finally, we have conducted an exhaustive study of network expressivness in terms of polychronous groups, based on various types of topologies (random, regular, small-world) and different parameters (number of neurones, conectivity, etc.). We are able to formulate some recommandations to create a network whose topology can be rich in terms of possible representations. We propose to link with the cognitive theory of multiple trace memory that can, in principle, be implemented and studied in the light of polychronous groups
Rosas, Flunger Rudolf. "L'approche de la dynamique des systèmes et l'aide à la décision multicritère comme outils d'apprentissage organisationnel". Paris 9, 1999. https://portail.bu.dauphine.fr/fileviewer/index.php?doc=1999PA090041.
Texto completoGaltier, Mathieu. "Une approche mathématique de l'apprentissage non-supervisé dans les réseaux de neurones récurrents". Phd thesis, École Nationale Supérieure des Mines de Paris, 2011. http://pastel.archives-ouvertes.fr/pastel-00667368.
Texto completoChaouche, Ahmed Chawki. "Une approche multi-agent pour la conception de systèmes d'intelligence ambiante : un modèle formel intégrant planification et apprentissage". Thesis, Paris 6, 2015. http://www.theses.fr/2015PA066084/document.
Texto completoThis work presents a concrete software architecture dedicated to ambient intelligence (AmI) features and requirements. The proposed behavioral model, called Higher-order Agent (HoA) captures the evolution of the mental representation of the agent and the one of its plan simultaneously. Plan expressions are written and composed using a formal algebraic language, namely AgLOTOS, so that plans are built automatically and on the fly, as a system of concurrent processes. Due to the compositional structure of AgLOTOS expressions, the updates of sub-plans are realized automatically accordingly to the revising of intentions, hence maintaining the consistency of the agent. Based on a specific semantics, a guidance service is also proposed to assist the agent in its execution. This guidance allows to improve the satisfaction of the agent's intentions with respect to the possible concurrent plans and the current context of the agent. Adopting the idea that "location" and "time" are key stones information in the activity of the agent, we show how to enforce guidance by ordering the different possible plans. As a major contribution, we demonstrate two original utility functions that are designed from the past-experiences of the action executions, and that can be combined accordingly to the current balance policy of the agent. A use case scenario is developed to show how the agent can act, even if it suffers from unexpected changes of contexts, it does not have many experiences and whose past experiences reveals some failure cases
Chaouche, Ahmed Chawki. "Une approche multi-agent pour la conception de systèmes d'intelligence ambiante : un modèle formel intégrant planification et apprentissage". Electronic Thesis or Diss., Paris 6, 2015. http://www.theses.fr/2015PA066084.
Texto completoThis work presents a concrete software architecture dedicated to ambient intelligence (AmI) features and requirements. The proposed behavioral model, called Higher-order Agent (HoA) captures the evolution of the mental representation of the agent and the one of its plan simultaneously. Plan expressions are written and composed using a formal algebraic language, namely AgLOTOS, so that plans are built automatically and on the fly, as a system of concurrent processes. Due to the compositional structure of AgLOTOS expressions, the updates of sub-plans are realized automatically accordingly to the revising of intentions, hence maintaining the consistency of the agent. Based on a specific semantics, a guidance service is also proposed to assist the agent in its execution. This guidance allows to improve the satisfaction of the agent's intentions with respect to the possible concurrent plans and the current context of the agent. Adopting the idea that "location" and "time" are key stones information in the activity of the agent, we show how to enforce guidance by ordering the different possible plans. As a major contribution, we demonstrate two original utility functions that are designed from the past-experiences of the action executions, and that can be combined accordingly to the current balance policy of the agent. A use case scenario is developed to show how the agent can act, even if it suffers from unexpected changes of contexts, it does not have many experiences and whose past experiences reveals some failure cases
Colliaux, David. "Classes de dynamiques neuronales et correlations structurées par l'experience dans le cortex visuel". Phd thesis, Ecole Polytechnique X, 2011. http://pastel.archives-ouvertes.fr/pastel-00676104.
Texto completoYin, Yuan. "Physics-Aware Deep Learning and Dynamical Systems : Hybrid Modeling and Generalization". Electronic Thesis or Diss., Sorbonne université, 2023. http://www.theses.fr/2023SORUS161.
Texto completoDeep learning has made significant progress in various fields and has emerged as a promising tool for modeling physical dynamical phenomena that exhibit highly nonlinear relationships. However, existing approaches are limited in their ability to make physically sound predictions due to the lack of prior knowledge and to handle real-world scenarios where data comes from multiple dynamics or is irregularly distributed in time and space. This thesis aims to overcome these limitations in the following directions: improving neural network-based dynamics modeling by leveraging physical models through hybrid modeling; extending the generalization power of dynamics models by learning commonalities from data of different dynamics to extrapolate to unseen systems; and handling free-form data and continuously predicting phenomena in time and space through continuous modeling. We highlight the versatility of deep learning techniques, and the proposed directions show promise for improving their accuracy and generalization power, paving the way for future research in new applications
Ayed, Ibrahim. "Neural Models for Learning Real World Dynamics and the Neural Dynamics of Learning". Electronic Thesis or Diss., Sorbonne université, 2022. http://www.theses.fr/2022SORUS434.
Texto completoThe work presented in this thesis was initially motivated by the discrepancy between the impressive performances of modern neural networks and the lack of applications to scientific problems for which data abounds. Focusing on evolution problems which are classically modelled through ordinary or partial differential equations~(O/PDEs) naturally brought us to consider the more general problem of representing and learning such equations from raw data with neural networks. This was the inception of the first part of our work. The point of view considered in this first part has a natural counterpart: what about the dynamics induced by the trajectories of the NN's weights during training or by the trajectories of data points within them during inference? Can they be usefully modelled? This question was the core of the second part of our work and, while theoretical tools other than O/PDEs happened to be useful in our analysis, our reasoning and intuition were fundamentally driven by considerations stemming from a dynamical viewpoint
Nguyen, Van Duong. "Variational deep learning for time series modelling and analysis : applications to dynamical system identification and maritime traffic anomaly detection". Thesis, Ecole nationale supérieure Mines-Télécom Atlantique Bretagne Pays de la Loire, 2020. http://www.theses.fr/2020IMTA0227.
Texto completoThis thesis work focuses on a class of unsupervised, probabilistic deep learning methods that use variational inference to create high capacity, scalable models for time series modelling and analysis. We present two classes of variational deep learning, then apply them to two specific problems related to the maritime domain. The first application is the identification of dynamical systems from noisy and partially observed data. We introduce a framework that merges classical data assimilation and modern deep learning to retrieve the differential equations that control the dynamics of the system. Using a state space formulation, the proposed framework embeds stochastic components to account for stochastic variabilities, model errors and reconstruction uncertainties. The second application is maritime traffic surveillance using AIS data. We propose a multitask probabilistic deep learning architecture can achieve state-of-the-art performance in different maritime traffic surveillance related tasks, such as trajectory reconstruction, vessel type identification and anomaly detection, while reducing significantly the amount data to be stored and the calculation time. For the most important task—anomaly detection, we introduce a geospatial detector that uses variational deep learning to builds a probabilistic representation of AIS trajectories, then detect anomalies by judging how likely this trajectory is
Caron, Alexandre. "Mesure de la dynamique des polluants gazeux en air intérieur : évaluation des performances de systèmes multi-capteurs". Thesis, Lille 1, 2016. http://www.theses.fr/2016LIL10161/document.
Texto completoNowadays, indoor air quality is a major health issue and a growing research challenge. Many pollutants are presentinside buildings. They are directly emitted by indoor sources such as building materials, furniture, occupants and theiractivities or transferred from outdoors. Due to an increasing concern for energy saving, recent buildings are much moreairtight, reducing the pollutants elimination to the outside. Standard analyzers are not suitable for monitoring the airquality indoors. These techniques are usually bulky, expensive, noisy and require skilled people. An alternative to theseconventional methods recently appeared under the form of microsensors. In this work, the performances and limitationsof different type of sensors such as infrared sensors, electrochemical sensors, photoionisation detectors orsemiconductive sensors for the measurement of CO2, CO, NOx, O3 or VOC, were evaluated in laboratory conditions andalso during measurement campaigns in order to monitor the major indoor air pollutants. Although the response of thesesensors is highly correlated with the concentration measured by reference instruments, their lack of selectivity does notalways allow a quantitative analysis. Naive Bayes classifier and bisecting k-means clustering were used to help analyzethe output of the sensors, and allow identifying typical pollution events, reflecting the dynamics of the indoor air quality
Gil, Quijano Javier. "Modèles d'auto-organisation pour l'émergence de formes urbaines à partir de comportements individuels à Bogota". Phd thesis, Université Pierre et Marie Curie - Paris VI, 2007. http://tel.archives-ouvertes.fr/tel-00270015.
Texto completoHartert, Laurent. "Reconnaissance des formes dans un environnement dynamique appliquée au diagnostic et au suivi des systèmes évolutifs". Phd thesis, Université de Reims - Champagne Ardenne, 2010. http://tel.archives-ouvertes.fr/tel-00549782.
Texto completoAmadou, Boubacar Habiboulaye. "Classification Dynamique de données non-stationnaires :Apprentissage et Suivi de Classes évolutives". Phd thesis, Université des Sciences et Technologie de Lille - Lille I, 2006. http://tel.archives-ouvertes.fr/tel-00106968.
Texto completoGuàrdia, Sebaoun Elie. "Accès personnalisé à l'information : prise en compte de la dynamique utilisateur". Thesis, Paris 6, 2017. http://www.theses.fr/2017PA066519/document.
Texto completoThe main goal of this thesis resides in using rich and efficient profiling to improve the adequation between the retrieved information and the user's expectations. We focus on exploiting as much feedback as we can (being clicks, ratings or written reviews) as well as context. In the meantime, the tremendous growth of ubiquitous computing forces us to rethink the role of information access platforms. Therefore, we took interest not solely in performances but also in accompanying users through their access to the information. Through this thesis, we focus on users dynamics modeling. Not only it improves the system performances but it also brings some kind of explicativity to the recommendation. Thus, we propose to accompany the user through his experience accessing information instead of constraining him to a given set of items the systems finds fitting
Guàrdia, Sebaoun Elie. "Accès personnalisé à l'information : prise en compte de la dynamique utilisateur". Electronic Thesis or Diss., Paris 6, 2017. http://www.theses.fr/2017PA066519.
Texto completoThe main goal of this thesis resides in using rich and efficient profiling to improve the adequation between the retrieved information and the user's expectations. We focus on exploiting as much feedback as we can (being clicks, ratings or written reviews) as well as context. In the meantime, the tremendous growth of ubiquitous computing forces us to rethink the role of information access platforms. Therefore, we took interest not solely in performances but also in accompanying users through their access to the information. Through this thesis, we focus on users dynamics modeling. Not only it improves the system performances but it also brings some kind of explicativity to the recommendation. Thus, we propose to accompany the user through his experience accessing information instead of constraining him to a given set of items the systems finds fitting
Lurette, Christophe. "Développement d'une technique neuronale auto-adaptative pour la classification dynamique de données évolutives : application à la supervision d'une presse hydraulique". Lille 1, 2003. https://ori-nuxeo.univ-lille1.fr/nuxeo/site/esupversions/aed48e38-323f-425b-b6ff-c8e75ff5d4b6.
Texto completoKusyk, Meryl. "Les dynamiques du développement de l'anglais au travers d'activités informelles en ligne : une étude exploratoire auprès d'étudiants français et allemands". Thesis, Strasbourg, 2017. http://www.theses.fr/2017STRAG037.
Texto completoPreliminary research regarding the online informal learning of English has shown that L2 development can result from participation in informal activities online. The goal of this dissertation is to examine the range of these online practices and to analyse university students’ long-term L2 development through their participation in such activities.953 French and German university students responded to a questionnaire containing approximately 60 questions regarding their online informal activities in English. Results from this survey show many similar practices between the two cohorts, a preference for comprehension over production and interaction activities, low rates of active (explicit) learning and content-associated rather than language-associated reasons for participating. Case studies were subsequently carried out. Oral and written data were collected over 10 months and analysed for complexity, accuracy and fluency measures as well as the use of language chunks. Results show that each language user interacts with the activities in his/her own unique style and that the different L2 measures evolve non-linearly and in relation to one another