Dissertations / Theses on the topic 'Modèles de neurones impulsionnels'
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 'Modèles de neurones impulsionnels.'
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.
Brette, Romain. "Modèles Impulsionnels de Réseaux de Neurones Biologiques." Phd thesis, Université Pierre et Marie Curie - Paris VI, 2003. http://tel.archives-ouvertes.fr/tel-00005340.
Rochel, Olivier. "Une approche événementielle pour la modélisation et la simulation de réseaux de neurones impulsionnels." Nancy 1, 2004. http://www.theses.fr/2004NAN10004.
At present, there exists no generic modeling and simulation framework that addresses the study of large spiking neural networks. In the existing models, the impulses are generally associated with discontinuities in the otherwise continuous dynamics of the neurons. This raises modeling and practical implementation issues. We propose an novel approach based on the discrete-event system abstraction, grounded on the DEVS formalism, that can be used to represent a large class of spiking neurons and permits the modeling of large networks. A simulation engine has been developed on top of this formalism. It is based on an efficient event-driven algorithm and has been evaluated on sequential as well as parallel machines. We have tested our approach within a multi-disciplinary project on olfactory perception
Quan, Zou. "Modèles computationnels de la plasticité impulsionnelle : synapses, neurones et circuits." Paris 6, 2006. http://www.theses.fr/2006PA066135.
Ambard, Maxime. "Influence de l'inhibition synaptique sur le codage de l'information par les cellules mitrales du bulbe olfactif." Phd thesis, Université Henri Poincaré - Nancy I, 2009. http://tel.archives-ouvertes.fr/tel-00401813.
Dans un premier temps, l'analyse de données expérimentales recueillies en condition in vitro dans des tranches de bulbe olfactif de rats révèle le caractère phasé des potentiels d'action des cellules mitrales relativement aux oscillations du potentiel de champ local. Ce phasage est largement atténué lorsque l'on bloque pharmacologiquement l'inhibition provenant des granules, mettant ainsi en évidence le rôle primordial de l'inhibition synaptique. Afin d'extraire le décours temporel de la conductance synaptique inhibitrice, nous proposons une nouvelle méthode basée sur l'ajustement d'un modèle de neurone associé à l'injection de bloqueurs synaptiques. Grâce à celle-ci, nous observons que les fluctuations de la conductance synaptique inhibitrice sont corrélées à celles mesurées sur le potentiel de champ local. Une relation entre l'inhibition reçue et la phase des potentiels d'action est également dévoilée. Un neurone aura plus de chance d'émettre en phase s'il reçoit un nombre important d'événements synaptiques inhibiteurs et si ces événements sont eux-même phasés.
Dans un deuxième temps, les résultats de cette analyse sont rassemblés au sein d'un modèle informatique de bulbe olfactif afin d'explorer les capacités de codage de l'interaction mitrale-granule. Après avoir montré que le transfert d'information des cellules mitrales semble plus résider dans leurs instants précis d'émission de potentiels d'action au cours des oscillations que dans leurs fréquences de décharges, nous étudions analytiquement l'influence du nombre d'événements synaptiques inhibiteurs reçus et de leur dispersion temporelle sur la précision de l'activité des cellules mitrales. Notre étude conclut que la robustesse du code produit par les cellules mitrales lors des oscillations du réseau est conditionnée par une forte interaction synaptique entre les cellules mitrales et les cellules granulaires. En dernier lieu, nous appliquons notre modèle de bulbe olfactif pour reconnaître des odeurs à l'aide d'une matrice de capteurs de gaz artificiels.
Chevallier, Sylvain. "Implémentation d'un système préattentionnel avec des neurones impulsionnels." Phd thesis, Université Paris Sud - Paris XI, 2009. http://tel.archives-ouvertes.fr/tel-00472849.
Oudjail, Veïs. "Réseaux de neurones impulsionnels appliqués à la vision par ordinateur." Electronic Thesis or Diss., Université de Lille (2022-....), 2022. http://www.theses.fr/2022ULILB048.
Artificial neural networks (ANN) have become a must-have technique in computer vision, a trend that started during the 2012 ImageNet challenge. However, this success comes with a non-negligible human cost for manual data labeling, very important in model learning, and a high energy cost caused by the need for large computational resources. Spiking Neural Networks (SNN) provide solutions to these problems. It is a particular class of ANNs, close to the biological model, in which neurons communicate asynchronously by representing information through spikes. The learning of SNNs can rely on an unsupervised rule: the STDP. It modulates the synaptic weights according to the local temporal correlations observed between the incoming and outgoing spikes. Different hardware architectures have been designed to exploit the properties of SNNs (asynchrony, sparse and local operation, etc.) in order to design low-power solutions, some of them dividing the cost by several orders of magnitude. SNNs are gaining popularity and there is growing interest in applying them to vision. Recent work shows that SNNs are maturing by being competitive with the state of the art on "simple" image datasets such as MNIST (handwritten numbers) but not on more complex datasets. However, SNNs can potentially stand out from ANNs in video processing. The first reason is that these models incorporate an additional temporal dimension. The second reason is that they lend themselves well to the use of event-driven cameras. They are bio-inspired sensors that perceive temporal contrasts in a scene, in other words, they are sensitive to motion. Each pixel can detect a light variation (positive or negative), which triggers an event. Coupling these cameras to neuromorphic chips allows the creation of totally asynchronous and massively parallelized vision systems. The objective of this thesis is to exploit the capabilities offered by SNNs in video processing. In order to explore the potential offered by SNNs, we are interested in motion analysis and more particularly in motion direction estimation. The goal is to develop a model capable of learning incrementally, without supervision and with few examples, to extract spatiotemporal features. We have therefore performed several studies examining the different points mentioned using synthetic event datasets. We show that the tuning of the SNN parameters is essential for the model to be able to extract useful features. We also show that the model is able to learn incrementally by presenting it with new classes without deteriorating the performance on the mastered classes. Finally, we discuss some limitations, especially on the weight learning, suggesting the possibility of more delay learning, which are still not very well exploited and which could mark a break with ANNs
Aziz, Mohammed, and Abdelaziz Bensrhair. "Apprentissage de réseaux de neurones impulsionnels. Application à des systèmes sensorimoteurs." INSA de Rouen, 2005. http://www.theses.fr/2005ISAM0005.
Lecerf, Gwendal. "Développement d'un réseau de neurones impulsionnels sur silicium à synapses memristives." Thesis, Bordeaux, 2014. http://www.theses.fr/2014BORD0219/document.
Supported financially by ANR MHANN project, this work proposes an architecture ofspiking neural network in order to recognize pictures, where traditional processing units are inefficient regarding this. In 2008, a new passive electrical component had been discovered : the memristor. Its resistance can be adjusted by applying a potential between its terminals. Behaving intrinsically as artificial synapses, memristives devices can be used inside artificial neural networks.We measure the variation in resistance of a ferroelectric memristor (obtained from UMjCNRS/Thalès) similar to the biological law STDP (Spike Timing Dependant Plasticity) used with spiking neurons. With our measurements on the memristor and our network simulation (aided by INRIASaclay) we designed successively two versions of the IC. The second IC design is driven by specifications of the first IC with additional functionalists. The second IC contains two layers of a spiking neural network dedicated to learn a picture of 81 pixels. A demonstrator of hybrid neural networks will be achieved by integrating a chip of memristive crossbar interfaced with thesecond IC
Boullet, Isabelle. "La sonie des sons impulsionnels : perception, mesures et modèles." Phd thesis, Université de la Méditerranée - Aix-Marseille II, 2005. http://tel.archives-ouvertes.fr/tel-00009870.
Boullet, Isabelle Catherine. "La sonie des sons impulsionnels : perception, mesures et modèles." Aix-Marseille 2, 2005. http://www.theses.fr/2005AIX22053.
Meunier, David. "UNE MODÉLISATION ÉVOLUTIONNISTE DU LIAGE TEMPOREL." Phd thesis, Université Lumière - Lyon II, 2007. http://tel.archives-ouvertes.fr/tel-00198797.
Nous avons développé un modèle de réseau de neurones impulsionnels, dont la topologie est modifiée par un algorithme évolutionniste. Le critère de performance utilisé pour l'algorithme évolutionniste est évalué par l'intermédiaire du comportement d'un individu contrôlé par un réseau de neurones impulsionnels, et placé dans un environnement virtuel. L'utilisation du neurone impulsionnel, ayant la propriété de détection de synchronie, oblige l'évolution à construire un système utilisant cette propriété au niveau global, d'où l'émergence de la synchronisation neuronale à large-échelle. Les propriétés topologiques et dynamiques du réseau de neurones ne sont pas prises en compte dans le calcul de la performance, mais sont étudiées a posteriori, en comparant les individus avant et après évolution.
D'une part, grâce aux outils de la théorie des réseaux complexes, nous montrons l'émergence d'un certain nombre de propriétés topologiques, notamment la propriété de réseau ``petit-monde''. Ces propriétés topologiques sont similaires à celles observées au niveau de l'anatomie des systèmes nerveux en biologie. D'autre part, au niveau de la dynamique, nous établissons que la propriété de synchronisation neuronale à large-échelle, résultant de la présentation d'un stimulus, est présente chez les individus évolués. Pour ce faire, nous nous appuyons sur les outils classiquement utilisés en électrophysiologie, et nous les étendons pour pouvoir interpréter la grande quantité de données obtenue à partir du modèle.
Le modèle montre que l'on peut construire des réseaux de neurones basés sur l'hypothèse du liage temporel en ayant recours à l'évolution artificielle, en se basant sur un critère de performance écologique, c.à.d. le comportement de l'individu dans son environnement. D'autre part, les outils développés pour l'analyse des propriétés du modèle peuvent être utilisés dans d'autres domaines, en premier lieu en électrophysiologie. En effet, à cause des progrès techniques sur les enregistrements électrophysiologiques, la quantité de données se rapproche singulièrement de celle issue du modèle.
Werner, Thilo. "Réseaux de neurones impulsionnels basés sur les mémoires résistives pour l'analyse de données neuronales." Thesis, Université Grenoble Alpes (ComUE), 2017. http://www.theses.fr/2017GREAS028/document.
The central nervous system of humankind is an astonishing information processing system in terms of its capabilities, versatility, adaptability and low energy consumption. Its complex structure consists of billions of neurons interconnected by trillions of synapses forming specialized clusters. Recently, mimicking those paradigms has attracted a strongly growing interest, triggered by the need for advanced computing approaches to tackle challenges related to the generation of massive amounts of complex data in the Internet of Things (IoT) era. This has led to a new research field, known as cognitive computing or neuromorphic engineering, which relies on the so-called non-von-Neumann architectures (brain-inspired) in contrary to von-Neumann architectures (conventional computers). In this thesis, we explore the use of resistive memory technologies such as oxide vacancy based random access memory (OxRAM) and conductive bridge RAM (CBRAM) for the design of artificial synapses that are a basic building block for neuromorphic networks. Moreover, we develop an artificial spiking neural network (SNN) based on OxRAM synapses dedicated to the analysis of spiking data recorded from the human brain with the goal of using the output of the SNN in a brain-computer interface (BCI) for the treatment of neurological disorders. The impact of reliability issues characteristic to OxRAM on the system performance is studied in detail and potential ways to mitigate penalties related to single device uncertainties are demonstrated. Besides the already well-known spike-timing-dependent plasticity (STDP) implementation with OxRAM and CBRAM which constitutes a form of long term plasticity (LTP), OxRAM devices were also used to mimic short term plasticity (STP). The fundamentally different functionalities of LTP and STP are put in evidence
Mouraud, Anthony. "Approche distribuée pour la simulation évènementielle de réseaux de neurones impulsionnels : application du contrôle des saccades oculaires." Antilles-Guyane, 2009. http://www.theses.fr/2009AGUY0271.
Simulating Spiking Neural Networks (SNN) with a sequentialevent-driven approach consumes less computation time than clock-drive methods. On the other hand, parallel computing supports provide a larger amount of material ressources for optimizing simulation performance. This PhD dissertation proposes DAMNED a Distributed And Multithreaded Neural Event-Driven simulation framework. DAMNED distributes the neurons and connections of the network on the material ressources synchronized through a decentralized globa virtual time and couples local multithreaded processing to the distributed hardware. DAMNED allows to speed up the simulation and to manage wider neural networks than sequential processing. DAMNED is suited to- run many models of spiking neurons and networks, and most material supports are workable. Using DAMNED is presented first on simple networks for different sizes, connectivities and activities. Next, DAMNED is applied to model and study the interactions between the neural circuits of the saccadic system located in the brainstem with SNN. The model helps validating the hypothesis that the saccade amplitude could be encoded by a vector summation of the activitie in the superior colliculus motor map rather than a vector average, compared to data obtained in the simulation. The originality of the present work is to couple event-driven and distributed programming: Moreover, DAMNED is the first SNN simulator taking advantage of an event-driven strategy internal multithreading of the logic processes and a distributed architecture of physical processes. Hence DAMNED is an advance in the area of simulating wide sizes of spiking neuron networks
Cherdo, Yann. "Détection d'anomalie non supervisée sur les séries temporelle à faible coût énergétique utilisant les SNNs." Electronic Thesis or Diss., Université Côte d'Azur, 2024. http://www.theses.fr/2024COAZ4018.
In the context of the predictive maintenance of the car manufacturer Renault, this thesis aims at providing low-power solutions for unsupervised anomaly detection on time-series. With the recent evolution of cars, more and more data are produced and need to be processed by machine learning algorithms. This processing can be performed in the cloud or directly at the edge inside the car. In such a case, network bandwidth, cloud services costs, data privacy management and data loss can be saved. Embedding a machine learning model inside a car is challenging as it requires frugal models due to memory and processing constraints. To this aim, we study the usage of spiking neural networks (SNNs) for anomaly detection, prediction and classification on time-series. SNNs models' performance and energy costs are evaluated in an edge scenario using generic hardware models that consider all calculation and memory costs. To leverage as much as possible the sparsity of SNNs, we propose a model with trainable sparse connections that consumes half the energy compared to its non-sparse version. This model is evaluated on anomaly detection public benchmarks, a real use-case of anomaly detection from Renault Alpine cars, weather forecasts and the google speech command dataset. We also compare its performance with other existing SNN and non-spiking models. We conclude that, for some use-cases, spiking models can provide state-of-the-art performance while consuming 2 to 8 times less energy. Yet, further studies should be undertaken to evaluate these models once embedded in a car. Inspired by neuroscience, we argue that other bio-inspired properties such as attention, sparsity, hierarchy or neural assemblies dynamics could be exploited to even get better energy efficiency and performance with spiking models. Finally, we end this thesis with an essay dealing with cognitive neuroscience, philosophy and artificial intelligence. Diving into conceptual difficulties linked to consciousness and considering the deterministic mechanisms of memory, we argue that consciousness and the self could be constitutively independent from memory. The aim of this essay is to question the nature of humans by contrast with the ones of machines and AI
Dumont, Grégory. "Analyse de modèles de population de neurones : cas des neurones à réponse postsynaptique par saut de potentiel." Thesis, Bordeaux 1, 2012. http://www.theses.fr/2012BOR14601/document.
This thesis concerns the mathematical modelling and the study of the behavior of a population of neurons. In this work we will mainly consider a population of excitatory neurons whe reall the cells of the network follow the integrate-and-fire model. Nonetheless, we will tackle in a chapter the modelling of an inhibitory population of neurons, and we will discuss in the lastchapter the modelling of a population of neurons that follows the Ermentrout-Koppell model.The point of view of this thesis is given by the population density approach that has beenintroduced more than a decade ago in order to facilitate the simulation of a large assembly ofneurons. More precisely, this approach gives a partial differential equation that describes thedensity of neurons in the state space that is the set of all admissible potential of a neuron. We will assume that when receiving an action potential, the potential of the neuron makes a small jump. As we will see this partial differential equation is non linear (due to the coupling betweenneurons) and non-local (due to the potential jump). If this idea is complicated and abstract, itallows to simulate easily a large neural network.First of all, the thesis gives a mathematical framework for the equations that arise from thisthe population density approach. Then we will discuss the existence and the possible blow upin finite time of the solution. We will discuss how the consideration of more realistic modellingassumptions, as the refractory period and the delay between the emission and the reception ofan action potential can stop the blow up of the solution and give a well posed model.We will also try to caracterise the occurence of synchronization of the neural network. Twodifferent ways of seeing the synchronization will be describe. One relates the blow up in finitetime of the solution to the occurence of a Dirac mass in the firing rate of the population.Nonetheless, taking into account the delays, this kind of blow up will not be observed anymore.Nonetheless, as we will see, with this additional features the model will generate some periodicalsolutions that can also be related to the synchronization of the population
Senneret, Marc. "Chaos et ergodicité pour une famille de modèles de neurones." Paris 7, 2007. http://www.theses.fr/2007PA077078.
This thesis present a mathematical analysis of models of neuronal activity. In a first part, we present the main results concerning the biology of neurons. We analyse two of the most used models of neuron, the Hodgkin-Huxley model and the FitzHugh-Nagumo model. . By a Poincaré section method, we make a simplier model, piece-wise linear, which keep the main features of excitability. We then study numericaly and analyticaly the dynamic of two coupled neurons, modeled by the precedent one. The second part is dedicated to the rigourous demonstration of the existence of invariant measures of SRB type for piece-wise affine maps of Rn, like our latter model. We use for this a method based on Frobenus-Perron operator and the inegality of Lasota-Yorke ; These results give the rigourous fondations for the results of the first part
Dupeyron, Denis. "Contribution à l'intégration sur silicium de modèles analogiques de neurones biologiques." Bordeaux 1, 1998. http://www.theses.fr/1998BOR10620.
Weng, Qilong. "Stabilité pour des modèles de réseaux de neurones et de chimiotaxie." Thesis, Paris Sciences et Lettres (ComUE), 2017. http://www.theses.fr/2017PSLED026/document.
This thesis is aimed to study some biological models in neuronal network and chemotaxis with the spectral analysis method. In order to deal with the main concerning problems, such as the existence and uniqueness of the solutions and steady states as well as the asymptotic behaviors, the associated linear or linearized model is considered from the aspect of spectrum and semigroups in appropriate spaces then the nonlinear stability follows. More precisely, we start with a linear runs-and-tumbles equation in dimension d≥1 to establish the existence of a unique positive and normalized steady state and the exponential asymptotic stability in weighted L¹ space based on the Krein-Rutman theory together with some moment estimates from kinetic theory. Then, we consider time elapsed model under general assumptions on the firing rate and prove the uniqueness of the steady state and its nonlinear exponential stability in case without or with delay in the weak connectivity regime from the spectral analysis theory for semigroups. Finally, we study the model under weaker regularity assumption on the firing rate and the existence of the solution as well as the same exponential stability are established generally no matter taking delay into account or not and no matter in weak or strong connectivity regime
Roth, Sophie. "Réseaux de neurones modèles : contrôle de la différenciation axonale par micropatterns." Phd thesis, Grenoble 1, 2009. http://www.theses.fr/2009GRE10262.
In vitro neuronal networks are pertinent simple systems to approach brain computational complexity. They are even more useful when an architecture evoking the in vivo network organization can be enforced. This is why micro-patterned substrates are now widely used to force cell adhesion and growth according to a predefined topology. However, building fully controlled neuronal microcircuits requires precise supervision of the information flow between cells, which can only be achieved by inducing neuronal polarity (i. E. Axonal differentiation) in a specific direction. Although more polarity-regulating molecules are being discovered each day, they are hardly usable to create polarized networks in vitro as most patterning technologies are not compatible with protein grafting. In this PhD, our goal was to achieve full control of neural polarity by combined action of non-specific adhesion and physical constraints provided by sophisticated patterning geometries. We report here the mastering of axonal growth direction with a success close to 90\%. This result was based upon previous observations : the centrosome localization determines the axon emergence point and mechanical tension is sufficient to ensure axon formation. We coupled these results into a single pattern to constrain the centrosome position with suitable adhesive patterns and prevent axon growing on undesired positions with specific curved shapes that provide a limitation of neurite tension. These findings not only provide an important tool for creating neuronal model networks but also question the centrosome function and the mechanisms of adhesion and force transmission within neurites that have been so far neglected in favour of growth cone analysis
Roth, Sophie. "Réseaux de Neurones modèles : Contrôle de la différenciation axonale par micropatterns." Phd thesis, Université de Grenoble, 2009. http://tel.archives-ouvertes.fr/tel-01010302.
Fois, Adrien. "Plasticité et codage temporel dans les réseaux impulsionnels appliqués à l'apprentissage de représentations." Electronic Thesis or Diss., Université de Lorraine, 2022. http://www.theses.fr/2022LORR0299.
Neuromorphic computing is a rapidly growing field of computer science. It seeks to define models of computation inspired by the properties of the brain. Neuromorphic computing redefines the nature of the three key components of learning: 1) data, 2) computing substrate, and 3) algorithms, based on how the brain works. First, the data are represented with all-or-nothing events distributed in space and time: spikes. Second, the computational substrate erases the separation between computation and memory introduced by Von Neumann architectures by co-locating them, as in the brain. Furthermore, the computation is massively parallel and asynchronous allowing the computational units to be activated on the fly, independently. Third, the learning algorithms are adapted to the computing substrate by exploiting the information available locally, at the neuron level. This vast overhaul in the way information transfer, information representation, computation and learning are approached, allows neuromorphic processors to promise in particular an energy saving of a considerable factor of 100 to 1000 compared to CPUs. In this thesis, we explore the algorithmic side of neuromorphic computing by proposing event-driven learning rules that satisfy locality constraints and are capable of extracting representations of event-based, sparse and asynchronous data streams. Moreover, while most related studies are based on rate codes where information is exclusively represented in the number of spikes, our learning rules exploit much more efficient temporal codes, where information is contained in the spike times. We first propose an in-depth analysis of a temporal coding method using a population of neurons. We propose a decoding method and we analyze the delivered information and the code structure. Then we introduce a new event-driven and local rule capable of extracting representations from temporal codes by storing centroids in a distributed way within the synaptic weights of a neural population. We then propose to learn representations not in synaptic weights, but rather in transmission delays operating intrinsically in the temporal dimension. This led to two new event-driven and local rules. One rule adapts delays so as to store representations, the other rule adapts weights so as to filter features according to their temporal variability. The two rules operate complementarily. In a last model, these rules adapting weights and delays are augmented by a new spatio-temporal neuromodulator. This neuromodulator makes it possible for the model to reproduce the behavior of self-organizing maps with spiking neurons, thus leading to the generation of ordered maps during the learning of representations. Finally, we propose a new generic labeling and voting method designed for spiking neural networks dealing with temporal codes. This method is used so as to evaluate our last model in the context of categorization tasks
Ferré, Paul. "Adéquation algorithme-architecture de réseaux de neurones à spikes pour les architectures matérielles massivement parallèles." Thesis, Toulouse 3, 2018. http://www.theses.fr/2018TOU30318/document.
The last decade has seen the re-emergence of machine learning methods based on formal neural networks under the name of deep learning. Although these methods have enabled a major breakthrough in machine learning, several obstacles to the possibility of industrializing these methods persist, notably the need to collect and label a very large amount of data as well as the computing power necessary to perform learning and inference with this type of neural network. In this thesis, we propose to study the adequacy between inference and learning algorithms derived from biological neural networks and massively parallel hardware architectures. We show with three contribution that such adequacy drastically accelerates computation times inherent to neural networks. In our first axis, we study the adequacy of the BCVision software engine developed by Brainchip SAS for GPU platforms. We also propose the introduction of a coarse-to-fine architecture based on complex cells. We show that GPU portage accelerates processing by a factor of seven, while the coarse-to-fine architecture reaches a factor of one thousand. The second contribution presents three algorithms for spike propagation adapted to parallel architectures. We study exhaustively the computational models of these algorithms, allowing the selection or design of the hardware system adapted to the parameters of the desired network. In our third axis we present a method to apply the Spike-Timing-Dependent-Plasticity rule to image data in order to learn visual representations in an unsupervised manner. We show that our approach allows the effective learning a hierarchy of representations relevant to image classification issues, while requiring ten times less data than other approaches in the literature
Tardif, Patrice. "Autostructuration des réseaux de neurones avec retards." Thesis, Université Laval, 2007. http://www.theses.ulaval.ca/2007/24240/24240.pdf.
Burel, Gilles. "RESEAUX DE NEURONES EN TRAITEMENT D'IMAGES - Des Modèles théoriques aux Applications Industrielles -." Phd thesis, Université de Bretagne occidentale - Brest, 1991. http://tel.archives-ouvertes.fr/tel-00101699.
traitement du signal et de l'image. On se place d'emblée du point de vue de
l'industriel impliqué dans la recherche, c'est à dire que l'on s'intéresse à
des problèmes réalistes, sans pour autant négliger la recherche
théorique.
Dans une première partie, nous montrons
l'intérêt des réseaux de neurones comme source d'inspiration pour la
conception de nouveaux algorithmes. Nous proposons en particulier une
structure originale pour la prédiction, ainsi que de nouveaux algorithmes de
Quantification Vectorielle. Les propriétés des algorithmes existants sont
également éclaircies du point de vue théorique, et des méthodes de réglage
automatique de leurs paramètres sont proposées.
On montre ensuite les capacités des réseaux de neurones à traiter un vaste champ
d'applications d'intérêt industriel. Pour divers problèmes de traitement de
l'image et du signal (de la segmentation à la séparation de sources, en
passant par la reconnaissance de formes et la compression de données), on
montre qu'il est possible de développer à moindre coût une solution neuronale
efficace.
Maghrebi, Fatine. "Modèles de réseaux de neurones pour la commande des carrefours à feux." Paris 1, 1994. http://www.theses.fr/1994PA010082.
Renversez, Gilles. "Modèles de neurones pour les neurosciences : des canaux ioniques à la synchronisation." Paris 11, 1997. http://www.theses.fr/1997PA112354.
We firstly recall the main characteristics and properties of biological neurons and we describe the Hodgkin-Huxley's model of neuronal electric activity. We then propose a model which allows to compute the spatiotemporal variations in the somatic transmembrane potential during an action potential. It takes into account bath the spatial distribution of ionic channels on a spherical body cell and their stochastic dynamics. This work has been motivated by recent experimental results which reported heterogeneous electric-field patterns on the somatic membrane during the action potential firing. We show that only inhomogeneous spatial distributions of ionic channels can create such heterogeneous patterns. Then, we describe a known method of dimensional reduction scheme which allows to link models using many variables to define neuronal dynamics to reduced systems. We study the dynamics of several such reduced models according to time-dependent external stimulation. In the following part, we study in detail the synchronization of two neurons mutually coupled with biologically plausible time-delayed inhibition. The neuron model is a two component dynamical system which can generate action potential. We obtain analytical results on the synchronization using piecewise linear approximations in neurons phase space. We show that the coincidence synchronization is stable versus the noise introduced in neuronal dynamics. Finally, we numerically study the dynamics of a globally connected network of neurons coupled with time-delayed inhibition. We use two types of quantitative measures to assess the level of synchrony of the network. We show that the synchronization of action potentials is stable versus noise even in an heterogeneous network and that the exchange of few action potentials is sufficient to get this synchronization. We notice that such a network of inhibitory neurons can generate membrane potential oscillations of the same type as the ones observed-in experiments
Burel, Gilles. "Réseaux de neurones en traitement d'images : des modèles théoriques aux applications industrielles." Brest, 1991. http://www.theses.fr/1991BRES2019.
Koiran, Pascal. "Puissance de calcul des réseaux de neurones artificiels." Lyon 1, 1993. http://www.theses.fr/1993LYO19003.
Yonaba, Harouna. "Modélisation hydrologique hybride : réseau de neurones - modèle conceptuel." Thesis, Université Laval, 2009. http://www.theses.ulaval.ca/2009/26583/26583.pdf.
Émirian, Frédéric. "Étude et conception d'une machine parallèle multi-modèles pour les réseaux de neurones." Toulouse, INPT, 1996. http://www.theses.fr/1996INPT091H.
Ducom, Jean-Christophe. "Codage temporel et apprentissage dans les réseaux de neurones." Aix-Marseille 1, 1996. http://www.theses.fr/1996AIX11041.
Benaïm, Michel. "Dynamiques d'activation et dynamiques d'apprentissage des réseaux de neurones." Toulouse, ENSAE, 1992. http://www.theses.fr/1992ESAE0001.
Ambroise, Matthieu. "Hybridation des réseaux de neurones : de la conception du réseau à l’interopérabilité des systèmes neuromorphiques." Thesis, Bordeaux, 2015. http://www.theses.fr/2015BORD0394/document.
HYBRID experiments allow to connect a biological neural network with an artificial one,used in neuroscience research and therapeutic purposes. During these three yearsof PhD, this thesis focused on hybridization in a close-up view (bi-diretionnal direct communication between the artificial and the living) and in a broader view (interoperability ofneuromorphic systems). In the early 2000s, an analog neuromorphic system has been connected to a biological neural network. This work is firstly, about the design of a digital neural network, for hybrid experimentsin two multi-disciplinary projects underway in AS2N team of IMS presented in this document : HYRENE (ANR 2010-Blan-031601), aiming the development of a hybrid system for therestoration of motor activity in the case of a spinal cord lesion,BRAINBOW (European project FP7-ICT-2011-C), aiming the development of innovativeneuro-prostheses that can restore communication around cortical lesions. Having a configurable architecture, a digital neural network was designed for these twoprojects. For the first project, the artificial neural network emulates the activity of CPGs (Central Pattern Generator), causing the locomotion in the animal kingdom. This activity will trigger aseries of stimuli in the injured spinal cord textit in vitro and recreating locomotion previously lost. In the second project, the neural network topology will be determined by the analysis anddecryption of biological signals from groups of neurons grown on electrodes, as well as modeling and simulations performed by our partners. The neural network will be able to repair the injuredneural network. This work show the two different networks design approach and preliminary results obtained in the two projects.Secondly, this work hybridization to extend the interoperability of neuromorphic systems. Through a communication protocol using Ethernet, it is possible to interconnect electronic neuralnetworks, computer and biological. In the near future, it will increase the complexity and size of networks
Langou, Karine. "Développement de nouveaux modèles expérimentaux de la Sclérose Latérale Amyotrophique." Thesis, Aix-Marseille 2, 2010. http://www.theses.fr/2010AIX22033.
ALS is a neurodegenerative disease characterized by a selective loss of motor neurons. A mutation in VAPB protein has been associated with ALS. VAPB, an endoplamic reticulum (ER) resident protein is proposed to play a role in protein transport and in the unfolded protein response. To manipulate VAPB (hVAPBwt and hVAPBp56s) expression in motor neurons in vitro, I used the viral gene transfer technology. hVAPBp56s induces selective motor neuron death which involved an ER-related pathway dependent on calcium signals. Studies on Cos-7 cells showed that hVAPBwt and hVAPBp56s impair the proteasome activity through the activation of ER stress and the sequestration of the 20S subnit. Moreover, we developed transgenic mice overxpressing hVAPBp56s which do not display any motor disorder
Gouix, Elsa. "Conséquences d'un dysfonctionnement astrocytaire sur la viabilité neuronale dans des modèles d'hypoxie/ischemie." Aix-Marseille 2, 2009. http://theses.univ-amu.fr.lama.univ-amu.fr/2009AIX22091.pdf.
Astrocytes have a crucial function in modulating neuronal activity especially in glutamatergic synapse where they tightly control transmission through sodium and secondarily ATP dependent-high affinity glutamate transporters. In hypoxic/ischemic (HI) conditions a major energetic crisis takes place and glutamate uptake has been shown to be stopped and aven reversed leading to extracellular glutamate concentration increase. My thesis results show that reverse glial glutamate uptake pharmacologically induced by PDC triggers neuronal death through extrasynaptic NMDA receptor induced calcium entry, neuronal mitochondrial membrane potential impairment and a shut off of the neurotrophic ERK 1/2 signaling pathway. This neuronal death wa rapid and necrotic. Using an oxygen and glucose deprivation model (OGD, HI model in vitro) we evidenced the oxidative and apoptotic death of differentiated (quiescent) murine striatal astrocytes 3 days after 3h-OGD. This death war correlated with a decreased capacity to synthetise glutathione the main antioxidant of the mamml CNS. By the same time, undifferentiated striatal astrocytes were resistant to 3h-OGD but changed transiently their morphology to Alzheimer type II astrocytes, a phenotype commonly observed in perinatal ischemic humain brain. Viability of striatal naïve neurons in coculture with differentiated astrocytes seemed to be higher than in coculture with undifferentiated ones. This could be explained by glutathione release from apoptotic astrocytes during the coculture, which can be neuroprotective. On the other hand a presumed glutamate release from undifferentiated astrocytes may activate extrasynaptic NMDA receptors and tigger neuronal death. These data illustrate that astrocytes may display different responses to HI insult thus conditioning neuronal
Schutz, Georges. "Adaptations et applications de modèles mixtes de réseaux de neurones à un processus industriel." Phd thesis, Université Henri Poincaré - Nancy I, 2006. http://tel.archives-ouvertes.fr/tel-00115770.
artificiels pour améliorer le contrôle de processus industriels
complexes, caractérisés en particulier par leur aspect temporel.
Les motivations principales pour traiter des séries temporelles
sont la réduction du volume de données, l'indexation pour la
recherche de similarités, la localisation de séquences,
l'extraction de connaissances (data mining) ou encore la
prédiction.
Le processus industriel choisi est un four à arc
électrique pour la production d'acier liquide au Luxembourg. Notre
approche est un concept de contrôle prédictif et se base sur des
méthodes d'apprentissage non-supervisé dans le but d'une
extraction de connaissances.
Notre méthode de codage se base sur
des formes primitives qui composent les signaux. Ces formes,
composant un alphabet de codage, sont extraites par une méthode
non-supervisée, les cartes auto-organisatrices de Kohonen (SOM).
Une méthode de validation des alphabets de codage accompagne
l'approche.
Un sujet important abordé durant ces recherches est
la similarité de séries temporelles. La méthode proposée est
non-supervisée et intègre la capacité de traiter des séquences de
tailles variées.
Suzano, Massa Francisco Vitor. "Mise en relation d'images et de modèles 3D avec des réseaux de neurones convolutifs." Thesis, Paris Est, 2017. http://www.theses.fr/2017PESC1198/document.
The recent availability of large catalogs of 3D models enables new possibilities for a 3D reasoning on photographs. This thesis investigates the use of convolutional neural networks (CNNs) for relating 3D objects to 2D images.We first introduce two contributions that are used throughout this thesis: an automatic memory reduction library for deep CNNs, and a study of CNN features for cross-domain matching. In the first one, we develop a library built on top of Torch7 which automatically reduces up to 91% of the memory requirements for deploying a deep CNN. As a second point, we study the effectiveness of various CNN features extracted from a pre-trained network in the case of images from different modalities (real or synthetic images). We show that despite the large cross-domain difference between rendered views and photographs, it is possible to use some of these features for instance retrieval, with possible applications to image-based rendering.There has been a recent use of CNNs for the task of object viewpoint estimation, sometimes with very different design choices. We present these approaches in an unified framework and we analyse the key factors that affect performance. We propose a joint training method that combines both detection and viewpoint estimation, which performs better than considering the viewpoint estimation separately. We also study the impact of the formulation of viewpoint estimation either as a discrete or a continuous task, we quantify the benefits of deeper architectures and we demonstrate that using synthetic data is beneficial. With all these elements combined, we improve over previous state-of-the-art results on the Pascal3D+ dataset by a approximately 5% of mean average viewpoint precision.In the instance retrieval study, the image of the object is given and the goal is to identify among a number of 3D models which object it is. We extend this work to object detection, where instead we are given a 3D model (or a set of 3D models) and we are asked to locate and align the model in the image. We show that simply using CNN features are not enough for this task, and we propose to learn a transformation that brings the features from the real images close to the features from the rendered views. We evaluate our approach both qualitatively and quantitatively on two standard datasets: the IKEAobject dataset, and a subset of the Pascal VOC 2012 dataset of the chair category, and we show state-of-the-art results on both of them
Asfogo, Noemi. "Intéraction entre synucléinopathie et dysfonctionnement mitochondrial dans des modèles neuronaux de la maladie de Parkinson." Electronic Thesis or Diss., Sorbonne université, 2023. https://accesdistant.sorbonne-universite.fr/login?url=https://theses-intra.sorbonne-universite.fr/2023SORUS732.pdf.
Parkinson's disease (PD) is a common neurodegenerative disorder characterized by the progressive loss of dopaminergic neurons in the substantia nigra pars compacta (SNpc) and the presence of inclusions called Lewy bodies (LBs), containing the presynaptic protein alpha-synuclein (αSyn) as a major component. A number of arguments, supported by advances in genetics, point to a role for altered αSyn homeostasis and mitochondrial dysfunction in PD. The αSyn species formed during its pathological accumulation are indeed suspected to affect the function of the mitochondria. However, we lack a consensual and integrated view of the consequences of these forms of αSyn on mitochondrial physiology over time, as well as the mechanisms involved. This thesis project was part of a detailed study aimed at exploring the contribution of αSyn-to mitochondrial dysfunction in neuronal and in vivo models, as part of the European IMI2/PD-MitoQUANT project. The thesis focused on the use of neuronal models of pathological αSyn aggregation, induced by exposure to Syn fibrils preformed in vitro (PFFs), to study the impact of this process on various aspects related to mitochondrial quality control. We first reproduced and validated in our laboratory a model of synucleinopathy based on exposure of primary mouse cortical neurons to PFFs. After two weeks of exposure, we observed the presence of deposits rich in phosphorylated αSyn (pS129) in these neurons, as well as a strong colocalization of pS129with mitochondria. We then assessed the impact of PFFs and synucleinopathy on mitophagy, using the fluorescent reporter MitoRosella. We demonstrated an increase in mitophagy following acute exposure of neurons to PFFs, which was not found in Parkin-deficient neurons (PRKN(-/-)). This suggests that PFFs damage mitochondria, activating a Parkin-dependent mitochondrial clearance program. A similar increase in mitophagy was found in neurons chronically exposed to PFFs in both wild type and PRKN(-/-) context, irrespective of the presence of pS129 deposits in the cells. In parallel, we monitored the impact of PFFs on mitochondrial biogenesis, studying de novo mitochondrial DNA (mtDNA) synthesis using an approach based on incorporation of the nucleoside analogue EdU and its visualization by imaging. We observed an increase in mtDNA synthesis following acute exposure to PFFs, which again did not occur in PRKN(-/-) neurons. In a context of chronic exposure, on the contrary, we observed a progressive decrease in de novo mtDNA synthesis, which was similar in the presence and absence of Parkin. In the last part of my thesis, I sought to confirm some of these key observations in a human neuronal model, and began exploring the mechanisms responsible for mitophagy activation in the synucleinopathy model. Taken together, these results demonstrate that progression of PFFs-induced Syn pathology is associated with imbalanced mitochondrial turnover, with on one hand enhanced mitochondrial degradation via mitophagy, and on the other hand lack of compensation via mitochondrial biogenesis. Ultimately, these alterations could lead to mitochondrial dysfunction, as shown by complementary results obtained in vitro and in vivo in the broader context of the PD-MitoQUANT project. Further studies are required to identify the mechanisms responsible for the observed alterations and determine to what extent they contribute to the neuronal vulnerability underlying neurodegeneration in PD
Zhu, Wenhua. "Modèles statistiques et réseaux de neurones : stratégie et validation dans le cas de la discrimination." Paris 9, 1995. https://portail.bu.dauphine.fr/fileviewer/index.php?doc=1995PA090022.
This thesis has accost two domains : data analysis and neural networks. It present the mains methods of discriminant analysis and treats the importants points in this field : learning procedures, construction and validation of decision rule, model selection, the relation with neural networks. This thesis propose some strategy of improving their generalization abilities and for reduce their complexity so that the neural networks can be applied to large more and realistic tasks
Bastien-Dionne, Pierre-Olivier. "Rôle de l'activité sensorielle dans la spécification du type cellulaire des neurones nouvellement générées dans le bulbe olfactif chez l'adulte." Master's thesis, Université Laval, 2008. http://hdl.handle.net/20.500.11794/20589.
Jeanblanc, Jérôme. "Etude des modalités d'implication des neurones dopaminergiques mésencéphaliques dans le phénomène d'inhibition latente." Université Louis Pasteur (Strasbourg) (1971-2008), 2003. http://www.theses.fr/2003STR13113.
The aim of this thesis was to study the ways in which mesencephalic dopaminergic (DA) neurons are involved in the latent inhibition (LI) phenomenon. This work formed part of a process of animal modelling of the physiopathology of schizophrenia. The importance of the work is demonstrated by the clinical data showing the existence of a disturbance of LI in non-treated schizophrenics. Neuropsychopharmacological data obtained over the last twenty years suggest that there is a dysfunction of the mesencephalic DA neurons in schizophrenia. To study the involvement of DA neurons in LI, we developed an original paradigm in a context of conditioned olfactory aversion. The release of DA was measured using in vivo voltammetry on freely moving rats. At first, the variations of the release of DA during the LI phenomenon were studied at the level of core, dorsal shell and ventral shell parts of the nucleus accumbens (Acc) and in the anterior and posterior subdivisions of the dorsal striatum. During the second part of the thesis, taking a physiopathological view, we tried to obtain a disturbance of LI. To this end, we studied to what extent functional blocking by tetrodotoxin of the entorhinal cortex (Ent) led to a disturbance of LI at the level of the behavioral answers and the DA answers obtained at the level of the core and dorsal shell parts of the Acc. We were able to establish that mesencephalic DA neurons innervating the core part and the dorsal shell part of the Acc and those innervating the anterior part of the dorsal striatum are involved in the LI phenomenon. The data obtained also made it possible to show that the Ent has a differential regulating effect at the level of the dorsal shell and core parts of the Acc. In conclusion, the work carried out during this thesis will contribute largely to the animal modelling of the disturbance of LI phenomenon observed in schizophrenics, and thus, ultimately, bring us closer to a model of the physiopathology of schizophrenia
Crépin, Sophie. "Comparaison de modèles murins de la maladie herpétique selon la souche virale et le site d'inoculation." Paris 11, 2009. http://www.theses.fr/2009PA114805.
Herpes simplex virus type 1 (HSV 1) is able to establish latent infection in neuronal tissues after a oral primary infection. Spontaneously or upon stimuli, HSV1 can reactive and induce ocular disease. The mechanisms involved in balance between latent infection and reactivation are not completely understood. In oro-ocular murin model of HSV1 (strain SC16) infection, the amounts of Latent-Associated Transcripts (LATs) and non-sliced ICPO transcripts, which encodes a protein involved in the reactivation process, have a similar pattern of evolution from the acute to the latent stage of infection, with an accumulation in trigeminal ganglia and a decrease in other latency sites. Such striking difference between types of neurones was partially found in the corneal scarification model of HSV1 infection. But, classical patterns of HSV1 latency, i. E. Complete extinction of early and late viral genes expression was not obtained in this model, especially with the KOS strain of HSV1. These results suggest that the biological patterns of HSV1 latency depend on the type of neurons, the viral strain and the murin model
Haddadi, Souad. "Réseaux de neurones, textures et modèles markoviens pour la détection et l'identification d'objets en mouvement." Compiègne, 1997. http://www.theses.fr/1997COMP1081.
In this PhD thesis, we present a method of analysis for image sequences. The method aims at dynamic scene interpretation where arbitrary objects evolve (in particular, human beings) and the scenes present non-uniform backgrounds and non-controlled illumination. Two processing approaches have been aborded : movement analysis (moving object detection) and pattern recognition (object identification). The proposed detection approach relies on a statistical segmentation procedure, which is based on the markovian principle and the analysis of texture. Considering an operator based on the differences between three successive images, taken two at a time, moving objects are detected, as well as the background regions which are discovered or occluded by these objects during their displacement. A coarse segmentation of this image operator is then applied to process the relevant zones of the image. This operation is then linked to a finer segmentation based on the markovian and textural principle. This problem was approached to a classification of the image operator into fixed and moving pixels. The identification approach of these objects uses another type of statistical model : the artificial neural networks, which allow computer training, after examples. Thus, models of neural network architectures were developed and applied to human being identification. The performances of these networks were calculated using two databases built for this project. We have demonstrated that high performances could be attained using MLP-type networks for our application. However, the studies accomplished during this thesis reveal a certain number of difficult problems. For example, in several cases we confronted the problem of selecting a pertinent training set
Augustin, Emmanuel. "Reconnaissance de mots manuscrits par systèmes hybrides : Réseaux de neurones et modèles de Markov cachés." Paris 5, 2001. http://www.theses.fr/2001PA05S026.
This thesis presents a recognition system for isolated handwritten words, given a dictionary, using a combination of neural networks (NN) and hidden markov models (HMM). NN and HMM have been extensively studied, the former in the field of isolated character recognition and the later in speech recognition, among other applications. Know-how on NN and HMM has motivated within the last 10 years many researches to combine the advantages of the two tools, that is discrimination power and sequence modelling. Some historical and original systems are recalled from speech and handwriting recognition. .
Chararas, Olivier. "Modèles connexionistes et application biomédicales : identification de souches bactériennes." Paris 5, 1993. http://www.theses.fr/1993PA05P026.
Mesci, Pinar. "L’implication du système xc- et du glutamate microglial dans les modèles murins de la sclérose latérale amyotrophique (SLA)." Paris 6, 2013. http://www.theses.fr/2013PA066286.
Amyotrophic lateral sclerosis (ALS) is the most common adult onset motor neuron disease leading to paralysis and death of patients. Mutations in SOD1 are responsible for motor neuron degeneration through a non-cell autonomous mechanism. Microglial cells, the macrophages of the central nervous system, participate in the progression of the disease. Since ALS is mainly sporadic, targeting the symptomatic phase during which microglial cells are actively involved is relevant to ALS. Since microglial neurotoxic factors are still largely unidentified in ALS and excitotoxicity is one pathway suggested to cause motor neuron death, our hypothesis was to assess if glutamate released by microglia through system xc- (a cystine/glutamate antiporter with the specific subunit xCT) could participate to motor neuron death in ALS. We now show that primary microglial cells expressed xCT and to a higher level upon activation, that xCT transcripts were enriched in microglia compared to the whole spinal cord and absent in motor neurons. In addition, xCT mRNA levels were increased in mutant SOD1 mouse spinal cords during disease progression. Deleting xCT in mutant SOD1 mice accelerated the onset of the disease but increased the duration of the symptomatic phase. Microglial system xc- was responsible for release of glutamate by microglial cells and deleting xCT increased the neurotrophic profile of microglial cells. These results show that system xc- could be a good target to slow ALS disease progression
Alexandre, Frédéric. "Une modélisation fonctionnelle du cortex : La colonne corticale : aspects visuels et moteurs." Nancy 1, 1990. http://docnum.univ-lorraine.fr/public/SCD_T_1990_0054_ALEXANDRE.pdf.
Meyer, Francisca. "Analyse des effets du blocage fonctionnel néonatal de plusieurs régions intégratives sur l’implication des neurones dopaminergiques striataux dans le phénomène d’inhibition latente." Strasbourg, 2009. http://www.theses.fr/2009STRA6201.
Rynkiewicz, Joseph. "Modèles hybrides intégrant des réseaux de neurones artificiels à des modèles de chaînes de Markov cachées : application à la prédiction de séries temporelles." Paris 1, 2000. http://www.theses.fr/2000PA010077.
Tay, Yong Haur. "Reconnaissance de l'écriture manuscrite hors-ligne par réseau de neurones artificiels et modèles de Markov cachés." Nantes, 2002. http://www.theses.fr/2002NANT2106.