Rozprawy doktorskie na temat „Réseaux de neurones sur graphes”
Utwórz poprawne odniesienie w stylach APA, MLA, Chicago, Harvard i wielu innych
Sprawdź 50 najlepszych rozpraw doktorskich naukowych na temat „Réseaux de neurones sur graphes”.
Przycisk „Dodaj do bibliografii” jest dostępny obok każdej pracy w bibliografii. Użyj go – a my automatycznie utworzymy odniesienie bibliograficzne do wybranej pracy w stylu cytowania, którego potrzebujesz: APA, MLA, Harvard, Chicago, Vancouver itp.
Możesz również pobrać pełny tekst publikacji naukowej w formacie „.pdf” i przeczytać adnotację do pracy online, jeśli odpowiednie parametry są dostępne w metadanych.
Przeglądaj rozprawy doktorskie z różnych dziedzin i twórz odpowiednie bibliografie.
Pasdeloup, Bastien. "Extending convolutional neural networks to irregular domains through graph inference". Thesis, Ecole nationale supérieure Mines-Télécom Atlantique Bretagne Pays de la Loire, 2017. http://www.theses.fr/2017IMTA0048/document.
Pełny tekst źródłaThis manuscript sums up our work on extending convolutional neuralnetworks to irregular domains through graph inference. It consists of three main chapters, each giving the details of a part of a methodology allowing the definition of such networks to process signals evolving on graphs with unknown structures.First, graph inference from data is explored, in order to provide a graph modeling the support of the signals to classify. Second, translation operators that preserve neighborhood properties of the vertices are identified on the inferred graph. Third, these translations are used to shift a convolutional kernel on the graph in order to define a convolutional neural network that is adapted to the input data.We have illustrated our methodology on a dataset of images. While not using any particular knowledge on the signals, we have been able to infer a graph that is close to a grid. Translations on this graph resemble Euclidean translations. Therefore, this has allowed us to define an adapted convolutional neural network that is very close what one would obtain when using the information that signals are images. This network, trained on the initial data, has out performed state of the art methods by more than 13 points, while using a very simple and easily improvable architecture.The method we have introduced is a generalization of convolutional neural networks. As a matter of fact, they can be seen as aparticularization of our approach in the case where the graph is a grid. Our work thus opens the way to numerous perspectives, as it provides an efficient way to build networks that are adapted to the data
Rosar, Kós Lassance Carlos Eduardo. "Graphs for deep learning representations". Thesis, Ecole nationale supérieure Mines-Télécom Atlantique Bretagne Pays de la Loire, 2020. http://www.theses.fr/2020IMTA0204.
Pełny tekst źródłaIn recent years, Deep Learning methods have achieved state of the art performance in a vast range of machine learning tasks, including image classification and multilingual automatic text translation. These architectures are trained to solve machine learning tasks in an end-to-end fashion. In order to reach top-tier performance, these architectures often require a very large number of trainable parameters. There are multiple undesirable consequences, and in order to tackle these issues, it is desired to be able to open the black boxes of deep learning architectures. Problematically, doing so is difficult due to the high dimensionality of representations and the stochasticity of the training process. In this thesis, we investigate these architectures by introducing a graph formalism based on the recent advances in Graph Signal Processing (GSP). Namely, we use graphs to represent the latent spaces of deep neural networks. We showcase that this graph formalism allows us to answer various questions including: ensuring generalization abilities, reducing the amount of arbitrary choices in the design of the learning process, improving robustness to small perturbations added to the inputs, and reducing computational complexity
Cattai, Tiziana. "Leveraging brain connectivity networks to detect mental states during motor imagery". Electronic Thesis or Diss., Sorbonne université, 2021. http://www.theses.fr/2021SORUS081.
Pełny tekst źródłaThe brain is a complex network and we know that inter-areal synchronization and de-synchronization mechanisms are crucial to perform motor and cognitive tasks. Nowadays, brain functional interactions are studied in brain-computer interface BCI) applications with more and more interest. This might have strong impact on BCI systems, typically based on univariate features which separately characterize brain regional activities. Indeed, brain connectivity features can be used to develop alternative BCIs in an effort to improve performance and to extend their real-life applicability. The ambition of this thesis is the investigation of brain functional connectivity networks during motor imagery (MI)-based BCI tasks. It aims to identify complex brain functioning, re-organization processes and time-varying dynamics, at both group and individual level. This thesis presents different developments that sequentially enrich an initially simple model in order to obtain a robust method for the study of functional connectivity networks. Experimental results on simulated and real EEG data recorded during BCI tasks prove that our proposed method well explains the variegate behaviour of brain EEG data. Specifically, it provides a characterization of brain functional mechanisms at group level, together with a measure of the separability of mental conditions at individual level. We also present a graph denoising procedure to filter data which simultaneously preserve the graph connectivity structure and enhance the signal-to-noise ratio. Since the use of a BCI system requires a dynamic interaction between user and machine, we finally propose a method to capture the evolution of time-varying data. In essence, this thesis presents a novel framework to grasp the complexity of graph functional connectivity during cognitive tasks
Corne, Christophe. "Parallélisation de réseaux de neurones sur architecture distribuée". Mulhouse, 1999. http://www.theses.fr/1999MULH0583.
Pełny tekst źródłaCarboni, Lucrezia. "Graphes pour l’exploration des réseaux de neurones artificiels et de la connectivité cérébrale humaine". Electronic Thesis or Diss., Université Grenoble Alpes, 2023. http://www.theses.fr/2023GRALM060.
Pełny tekst źródłaThe main objective of this thesis is to explore brain and artificial neural network connectivity from agraph-based perspective. While structural and functional connectivity analysis has been extensivelystudied in the context of the human brain, there is a lack of a similar analysis framework in artificialsystems.To address this gap, this research focuses on two main axes.In the first axis, the main objective is to determine a healthy signature characterization of the humanbrain resting state functional connectivity. To achieve this objective, a novel framework is proposed,integrating traditional graph statistics and network reduction tools, to determine healthy connectivitypatterns. Hence, we build a graph pair-wise comparison and a classifier to identify pathological statesand rank associated perturbed brain regions. Additionally, the generalization and robustness of theproposed framework were investigated across multiple datasets and variations in data quality.The second research axis explores the benefits of brain-inspired connectivity exploration of artificialneural networks (ANNs) in the future perspective of more robust artificial systems development. Amajor robustness issue in ANN models is represented by catastrophic forgetting when the networkdramatically forgets previously learned tasks when adapting to new ones. Our work demonstrates thatgraph modeling offers a simple and elegant framework for investigating ANNs, comparing differentlearning strategies, and detecting deleterious behaviors such as catastrophic forgetting.Moreover, we explore the potential of leveraging graph-based insights to effectively mitigatecatastrophic forgetting, laying a foundation for future research and explorations in this area
Alvado, Ludovic. "Neurones artificiels sur silicium : une évolution vers les réseaux". Bordeaux 1, 2003. http://www.theses.fr/2003BOR12674.
Pełny tekst źródłaThis thesis describes a new approach for modelling biological neuron networks. This approach uses analogue specific integrated circuit (ASIC) in which Hodgkin-Huxley formalism as been implemented to integrate medium density artificial neural network, modelled at a biological realistic level. This thesis also deals with the component mismatches problem and the pertinent choice of optimized structure dedicated to network applications
Bissery, Christophe. "La détection centralisée des fuites sur les réseaux d'eau potable par réseaux de neurones". Lyon, INSA, 1994. http://www.theses.fr/1994ISAL0112.
Pełny tekst źródłaFor few years, under the influence of the urban environment, the perception of dysfunction risk in technical systems and in particular in water supply networks has changed. The lack of risk doesn't exist and it's necessary to learn how to manage it. It's in this way that appears the need of centralized leakage detection on water supply networks, leaks that represent an important part of the dysfunction risk of water supply. This study proposes a centralized leakage detection system using a computerized neural network approach. The building method of learning bases and the sensors localization method are pointed out and developed. This study has showed that on a realistic network model results obtained with the centralized leakage detection system using a computerized neural network approach allowed experimentations on real networks. The study ends on the presentation of the working priorities for these real experimentations (and in particular the need of hourly water consumption previsions)
Faÿ, Armelle. "Sur la propagation de l'information dans les réseaux probabilistes". Paris 6, 1997. http://www.theses.fr/1997PA066770.
Pełny tekst źródłaWang, Shengrui. "Réseaux multicouches de neurones artificiels : algorithmes d'apprentissage, implantations sur hypercube : applications". Phd thesis, Grenoble INPG, 1989. http://tel.archives-ouvertes.fr/tel-00335818.
Pełny tekst źródłaLaflaquière, Arnaud. "Neurones artificiels sur silicium : conception analogique et construction de réseaux hybrides". Bordeaux 1, 1998. http://www.theses.fr/1998BOR10617.
Pełny tekst źródłaHaji, Mirsadeghi Mir Omid. "Routage sur les graphes géométriques aléatoires". Paris 6, 2012. http://www.theses.fr/2012PA066204.
Pełny tekst źródłaThe two first chapters are focused on preliminaries. In Chapter 3 of this thesis, we analyze a class of “Signal to Interference and Noise Ratio” (SINR) random graphs. These random graphs arise in the modeling of packet transmissions in wireless networks. In contrast to previous studies on SINR graphs, we consider both a space and a time dimension. We study optimal paths in such wireless networks in terms of first passage percolation on this random graph. We establish both positive and negative results on the associated time constant. The main negative result states that this time constant is infinite on the random graph associated with a Poisson point process under natural assumptions on the wireless channels. The main positive result states that when adding a periodic node infrastructure of arbitrarily small intensity to the Poisson point process, the time constant is positive and finite. In the second part, we develop a framework for studying point-map invariant measures. We focus on the case of a not necessarily bijective point-map f. We introduce the notion of Point-map Palm version of the point process Φ, which satisfies the desired invariance property when it exists and we give sufficient conditions for it to exist. Chapter 5, explains the connection between Chapters 3 and 4. It generalizes the notion of point-map Palm measures for stochastic point-maps and time dependent point-maps. As we will see in the end of the Chapter 3, the optimal path in the time- space SINR graph is not computable locally in time. This fact leads us to considering suboptimal local algorithms
Albano, Alice. "Dynamique des graphes de terrain : analyse en temps intrinsèque". Thesis, Paris 6, 2014. http://www.theses.fr/2014PA066260/document.
Pełny tekst źródłaWe are surrounded by a multitude of interaction networks from different contexts. These networks can be modeled as graphs, called complex networks. They have a community structure, i.e. groups of nodes closely related to each other and less connected with the rest of the graph. An other phenomenon studied in complex networks in many contexts is diffusion. The spread of a disease is an example of diffusion. These phenomena are dynamic and depend on an important parameter, which is often little studied: the time scale in which they are observed. According to the chosen scale, the graph dynamics can vary significantly. In this thesis, we propose to study dynamic processes using a suitable time scale. We consider a notion of relative time which we call intrinsic time, opposed to "traditional" time, which we call extrinsic time. We first study diffusion phenomena using intrinsic time, and we compare our results with an extrinsic time scale. This allows us to highlight the fact that the same phenomenon observed at two different time scales can have a very different behavior. We then analyze the relevance of the use of intrinsic time scale for detecting dynamic communities. Comparing communities obtained according extrinsic and intrinsic scales shows that the intrinsic time scale allows a more significant detection than extrinsic time scale
Albano, Alice. "Dynamique des graphes de terrain : analyse en temps intrinsèque". Electronic Thesis or Diss., Paris 6, 2014. http://www.theses.fr/2014PA066260.
Pełny tekst źródłaWe are surrounded by a multitude of interaction networks from different contexts. These networks can be modeled as graphs, called complex networks. They have a community structure, i.e. groups of nodes closely related to each other and less connected with the rest of the graph. An other phenomenon studied in complex networks in many contexts is diffusion. The spread of a disease is an example of diffusion. These phenomena are dynamic and depend on an important parameter, which is often little studied: the time scale in which they are observed. According to the chosen scale, the graph dynamics can vary significantly. In this thesis, we propose to study dynamic processes using a suitable time scale. We consider a notion of relative time which we call intrinsic time, opposed to "traditional" time, which we call extrinsic time. We first study diffusion phenomena using intrinsic time, and we compare our results with an extrinsic time scale. This allows us to highlight the fact that the same phenomenon observed at two different time scales can have a very different behavior. We then analyze the relevance of the use of intrinsic time scale for detecting dynamic communities. Comparing communities obtained according extrinsic and intrinsic scales shows that the intrinsic time scale allows a more significant detection than extrinsic time scale
Seube, Nicolas. "Régulation de systèmes contrôlés avec contraintes sur l'état par réseaux de neurones". Paris 9, 1992. https://portail.bu.dauphine.fr/fileviewer/index.php?doc=1992PA090013.
Pełny tekst źródłaCherif, Makrem. "Optimisation de l'ordonnancement par l'approche hybride basée sur les réseaux de neurones". Mémoire, École de technologie supérieure, 2004. http://espace.etsmtl.ca/712/1/CHERIF_Makrem.pdf.
Pełny tekst źródłaFrydlender, Hervé. "Implantation de réseaux de neurones artificiels sur multi-processeurs à mémoire distribuée". Grenoble INPG, 1992. http://www.theses.fr/1992INPG0132.
Pełny tekst źródłaHérault, Laurent. "Réseaux de neurones récursifs pour l'optimisation combinatoire : application à la théorie des graphes et à la vision par ordinateur". Grenoble INPG, 1991. http://www.theses.fr/1991INPG0019.
Pełny tekst źródłaIssartel, Yann. "Inférence sur des graphes aléatoires". Thesis, université Paris-Saclay, 2020. http://www.theses.fr/2020UPASM019.
Pełny tekst źródłaThis thesis lies at the intersection of the theories of non-parametric statistics and statistical learning. Its goal is to provide an understanding of statistical problems in latent space random graphs. Latent space models have emerged as useful probabilistic tools for modeling large networks in various fields such as biology, marketing or social sciences. We first define an identifiable index of the dimension of the latent space and then a consistent estimator of this index. More generally, such identifiable and interpretable quantities alleviate the absence of identifiability of the latent space itself. We then introduce the pair-matching problem. From a non-observed graph, a strategy sequentially queries pairs of nodes and observes the presence/absence of edges. Its goal is to discover as many edges as possible with a fixed budget of queries. For this bandit type problem, we study optimal regrets in stochastic block models and random geometric graphs. Finally, we are interested in estimating the positions of the nodes in the latent space, in the particular situation where the space is a circle in the Euclidean plane. For each of the three problems, we obtain procedures that achieve the statistical optimal performance, as well as efficient procedures with theoretical guarantees. These algorithms are analysed from a non-asymptotic viewpoint, relying in particular on concentration inequalities
Poirier, Carl. "Assemblage d'ADN avec graphes de de Bruijn sur FPGA". Master's thesis, Université Laval, 2015. http://hdl.handle.net/20.500.11794/27132.
Pełny tekst źródłaLécuyer, Fabrice. "Ordonner les nœuds pour passer à l'échelle sur les grands réseaux réels". Electronic Thesis or Diss., Sorbonne université, 2023. http://www.theses.fr/2023SORUS172.
Pełny tekst źródłaThis thesis focuses on using theoretical tools of computer science to improve algorithms in practice, specifically algorithms that process data in the form of graphs. A graph represents elements (nodes) and their interactions (edges). Computer scientists have designed theoretical algorithms for arbitrary graphs, such as finding shortest paths or identifying inter-connected nodes. However, real-world networks have specific properties that are unknown in advance due to the situations from which they arise. They can be very large, which presents a challenge for processing them in reasonable time. To help design scalable algorithms for real-world networks, we focus on the technique of node ordering, which consists in processing the nodes in a specific order that depends on local or global properties of the network. We provide a review on the different mechanisms and methods that have been used to design orderings across various application domains. Then, we present three contributions that use node orderings to make algorithms more efficient. First, we replicate a paper that designs an ordering to make cache systems more effective, which accelerates different graph algorithms. Second, we create new orderings that diminish the number of operations in an existing algorithm for triangle listing. Third, we use greedy algorithms with certain orderings to bound the size of a minimum vertex cover on a specific instance, which allows us to certify the quality of approximate values. These findings insist on scalability issues, time measurements, mathematical grounding and validation by experiments. Finally, we present a collaboration on network analysis that consists in describing the mobility of researchers within the space of knowledge
Ateme-Nguema, Barthélemy H. "Conception optimale des cellules de fabrication flexibles basée sur l'approche par réseaux de neurones". Mémoire, École de technologie supérieure, 2007. http://espace.etsmtl.ca/548/1/ATEME%2DNGUEMA_Barth%C3%A9lemy_Hugues.pdf.
Pełny tekst źródłaWerner, 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.
Pełny tekst źródłaThe 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
Allard, Antoine. "Percolation sur graphes aléatoires - modélisation et description analytique -". Thesis, Université Laval, 2014. http://www.theses.ulaval.ca/2014/30822/30822.pdf.
Pełny tekst źródłaGraphs are abstract mathematical objects used to model the interactions between the elements of complex systems. Their use is motivated by the fact that there exists a fundamental relationship between the structure of these interactions and the macroscopic properties of these systems. The structure of these graphs is analyzed within the paradigm of percolation theory whose tools and concepts contribute to a better understanding of the conditions for which these emergent properties appear. The underlying interactions of a wide variety of complex systems share many universal structural properties, and including these properties in a unified theoretical framework is one of the main challenges of the science of complex systems. Capitalizing on a multitype approach, a simple yet powerful idea, we have unified the models of percolation on random graphs published to this day in a single framework, hence yielding the most general and realistic framework to date. More than a mere compilation, this framework significantly increases the structural complexity of the graphs that can now be mathematically handled, and, as such, opens the way to many new research opportunities. We illustrate this assertion by using our framework to validate hypotheses hinted at by empirical results. First, we investigate how the network structure of some complex systems (e.g., power grids, social networks) enhances our ability to monitor them, and ultimately to control them. Second, we test the hypothesis that the “k-core” decomposition can act as an effective structure of graphs extracted from real complex systems. Third, we use our framework to identify the conditions for which a new immunization strategy against infectious diseases is optimal.
Comin, Carlo. "Complexité dans les Jeux Infinis sur les Graphes et les Réseaux de Contraintes Temporelles". Thesis, Paris Est, 2017. http://www.theses.fr/2017PESC1061/document.
Pełny tekst źródłaThis dissertation deals with a number of algorithmic problems motivated by automated temporal planning and formal verification of reactive and finite state systems. We focused on game theoretical methods to obtain novel insights, improved complexity bounds, and faster algorithms for the following models: Hyper Temporal Networks, Conditional Simple/Hyper Temporal Networks, Update Games, Muller McNaughton Games, and Mean Payoff Games
Vialatte, Jean-Charles. "Convolution et apprentissage profond sur graphes". Thesis, Ecole nationale supérieure Mines-Télécom Atlantique Bretagne Pays de la Loire, 2018. http://www.theses.fr/2018IMTA0118/document.
Pełny tekst źródłaConvolutional neural networks have proven to be the deep learning model that performs best on regularly structured datasets like images or sounds. However, they cannot be applied on datasets with an irregular structure (e.g. sensor networks, citation networks, MRIs). In this thesis, we develop an algebraic theory of convolutions on irregular domains. We construct a family of convolutions that are based on group actions (or, more generally, groupoid actions) that acts on the vertex domain and that have properties that depend on the edges. With the help of these convolutions, we propose extensions of convolutional neural netowrks to graph domains. Our researches lead us to propose a generic formulation of the propagation between layers, that we call the neural contraction. From this formulation, we derive many novel neural network models that can be applied on irregular domains. Through benchmarks and experiments, we show that they attain state-of-the-art performances, and beat them in some cases
Eiche, Antoine. "Ordonnancement temps réel pour architectures hétérogènes reconfigurables basé sur des structures de réseaux de neurones". Phd thesis, Université Rennes 1, 2012. http://tel.archives-ouvertes.fr/tel-00783893.
Pełny tekst źródłaBelley, Katia. "Détection de sites sécuritaires par réseaux de neurones pour un atterrissage autonome sur corps planétaire". Mémoire, Université de Sherbrooke, 2008. http://savoirs.usherbrooke.ca/handle/11143/1447.
Pełny tekst źródłaPaugam-Moisy, Hélène. "Optimisation des réseaux de neurones artificiels : analyse et mises en œuvre sur ordinateurs massivement parallèles". Lyon 1, 1992. http://www.theses.fr/1992LYO10018.
Pełny tekst źródłaVidal, Martin. "Architecture systolique pour un algorithme basé sur les réseaux de neurones pour l'égalisation de canaux". Thèse, Université du Québec à Trois-Rivières, 1999. http://depot-e.uqtr.ca/3409/1/000662425.pdf.
Pełny tekst źródłaTremblay, Nicolas. "Réseaux et signal : des outils de traitement du signal pour l'analyse des réseaux". Thesis, Lyon, École normale supérieure, 2014. http://www.theses.fr/2014ENSL0938/document.
Pełny tekst źródłaThis thesis describes new tools specifically designed for the analysis of networks such as social, transportation, neuronal, protein, communication networks... These networks, along with the rapid expansion of electronic, IT and mobile technologies are increasingly monitored and measured. Adapted tools of analysis are therefore very much in demand, which need to be universal, powerful, and precise enough to be able to extract useful information from very different possibly large networks. To this end, a large community of researchers from various disciplines have concentrated their efforts on the analysis of graphs, well define mathematical tools modeling the interconnected structure of networks. Among all the considered directions of research, graph signal processing brings a new and promising vision : a signal is no longer defined on a regular n-dimensional topology, but on a particular topology defined by the graph. To apply these new ideas on the practical problems of network analysis paves the way to an analysis firmly rooted in signal processing theory. It is precisely this frontier between signal processing and network science that we explore throughout this thesis, as shown by two of its major contributions. Firstly, a multiscale version of community detection in networks is proposed, based on the recent definition of graph wavelets. Then, a network-adapted bootstrap method is introduced, that enables statistical estimation based on carefully designed graph resampling schemes
Hafidi, Hakim. "Robust machine learning for Graphs/Networks". Electronic Thesis or Diss., Institut polytechnique de Paris, 2023. http://www.theses.fr/2023IPPAT004.
Pełny tekst źródłaThis thesis addresses advancements in graph representation learning, focusing on the challengesand opportunities presented by Graph Neural Networks (GNNs). It highlights the significanceof graphs in representing complex systems and the necessity of learning node embeddings that capture both node features and graph structure. The study identifies key issues in GNNs, such as their dependence on high-quality labeled data, inconsistent performanceacross various datasets, and susceptibility to adversarial attacks.To tackle these challenges, the thesis introduces several innovative approaches. Firstly, it employs contrastive learning for node representation, enabling self-supervised learning that reduces reliance on labeled data. Secondly, a Bayesian-based classifier isproposed for node classification, which considers the graph’s structure to enhance accuracy. Lastly, the thesis addresses the vulnerability of GNNs to adversarialattacks by assessing the robustness of the proposed classifier and introducing effective defense mechanisms.These contributions aim to improve both the performance and resilience of GNNs in graph representation learning
Galerne, Pascal. "Détection et classification de cibles posées sur le fond marin par réseaux de neurones en imagerie sonar". Brest, 1998. http://www.theses.fr/1998BRES2022.
Pełny tekst źródłaDumas, Maxime. "AlertWheel visualisation radiale de graphes bipartis appliquée aux systèmes de détection d'intrusions sur des réseaux informatiques". Mémoire, École de technologie supérieure, 2011. http://espace.etsmtl.ca/959/1/DUMAS_Maxime.pdf.
Pełny tekst źródłaGuyot, Dimitri. "Evaluation sur modèle de simulation thermique dynamique calibré des performances d’un contrôleur prédictif basé sur l’utilisation de réseaux de neurones". Thesis, Paris, HESAM, 2020. http://www.theses.fr/2020HESAC022.
Pełny tekst źródłaThe development of machine learning techniques, particularly neural networks, combined with the development of new information and communication technologies, is shaking up our societies through technological advances in a variety of sectors. The building sector is not spared, so these techniques may represent an interesting opportunity in a context where greenhouse gas emissions must be drastically reduced. The objective of this work is to assess the interest of these techniques in the field of building energy, with the aim of reducing energy consumption and improving thermal comfort. In addition, we ensure that this evaluation is carried out with a global vision, by placing the possible advantages in front of the different needs relating to the development of these technologies. This thesis work is organized in three parts preceded by a detailed introduction intended to give the reader an overview of the various contextual elements, thus allowing the thesis work to be placed in perspective. We then give in the first part the theoretical framework needed to understand the problems encountered during the elaboration and creation of neural networks for building energy applications. Then, a bibliographical study giving the reader a broad overview of the various applications of neural networks in the field of building energy is presented. The second part is devoted to the calibration of the building model that is then used to test and evaluate a predictive controller implementing neural networks. After an explanation of the method used and a detailed presentation of the model, a complete analysis of the calibration results is carried out. We conclude this part with observations and recommendations regarding the standard calibration guidelines recommended by three international organizations. Finally, a practical application using neural networks for the predictive control of indoor temperature is presented in the third part. After a theoretical introduction concerning predictive control, we detail the method employed to train the neural networks used. The results obtained in simulation with a predictive controller are then analyzed and compared with those obtained with two reference controllers for various simulation hypothesis. The predictive controller is thus tested in several scenarios, ranging from an ideal situation to more realistic operating conditions, including two different types of heat emitters, namely radiant ceilings and underfloor heating
Abadi, Mehdi. "Réalisation d'un réseau de neurones "SOM" sur une architecture matérielle adaptable et extensible à base de réseaux sur puce "NoC"". Thesis, Université de Lorraine, 2018. http://www.theses.fr/2018LORR0068/document.
Pełny tekst źródłaSince its introduction in 1982, Kohonen’s Self-Organizing Map (SOM) showed its ability to classify and visualize multidimensional data in various application fields. Hardware implementations of SOM, by exploiting the inherent parallelism of the Kohonen algorithm, allow to increase the overall performances of this neuronal network, often at the expense of the flexibility. On the other hand, the flexibility is offered by software implementations which on their side are not suited for real-time applications due to the limited time performances. In this thesis we proposed a distributed, adaptable, flexible and scalable hardware architecture of SOM based on Network-on-Chip (NoC) designed for FPGA implementation. Moreover, based on this approach we also proposed a novel hardware architecture of a growing SOM able to evolve its own structure during the learning phase
Abadi, Mehdi. "Réalisation d'un réseau de neurones "SOM" sur une architecture matérielle adaptable et extensible à base de réseaux sur puce "NoC"". Electronic Thesis or Diss., Université de Lorraine, 2018. http://www.theses.fr/2018LORR0068.
Pełny tekst źródłaSince its introduction in 1982, Kohonen’s Self-Organizing Map (SOM) showed its ability to classify and visualize multidimensional data in various application fields. Hardware implementations of SOM, by exploiting the inherent parallelism of the Kohonen algorithm, allow to increase the overall performances of this neuronal network, often at the expense of the flexibility. On the other hand, the flexibility is offered by software implementations which on their side are not suited for real-time applications due to the limited time performances. In this thesis we proposed a distributed, adaptable, flexible and scalable hardware architecture of SOM based on Network-on-Chip (NoC) designed for FPGA implementation. Moreover, based on this approach we also proposed a novel hardware architecture of a growing SOM able to evolve its own structure during the learning phase
Benasser, Ahmer. "L'accessibilité dans les réseaux de Pétri : une approche basée sur la programmation par contraintes". Lille 1, 2000. https://pepite-depot.univ-lille.fr/RESTREINT/Th_Num/2000/50376-2000-65.pdf.
Pełny tekst źródłaLa complexité de l'exploration du graphe d'accessibilité est alors repoussée au niveau de la résolution de contraintes : la propagation des contraintes nous interdit d'explorer les branches qui correspondent à des séquences de steps qui ne mènent pas au marquage final désiré. Pour résoudre les problèmes d'ordonnancement, il est nécessaire d'introduire l'aspect temporel. Nous définissons un modèle de réseau de Pétri temporisé et autonome. A chaque marquage, nous associons une date pour chacune des places. Cette date correspond à la date de création du dernier jeton dans cette place. Ainsi, le temps n'est pas contrôlée par une horloge externe. Ce sont les tirs des transitions qui font varier localement au niveau des places le temps. L'algorithme d'accessibilité peut alors être adapté pour ce type de réseau. Nous obtenons alors des séquences de tirs dates qui peuvent etre interprétés comme des ordonnancements réalisables
Benaouda, Djamel. "Modélisation et simulation d'un réseau de neurones formels : implantation sur machine parallèle "hypercube FPS T-40". Phd thesis, Grenoble 1, 1992. http://tel.archives-ouvertes.fr/tel-00340978.
Pełny tekst źródłaLe, Gal Jean-Patrick. "Coordination locomotion-respiration : influences des réseaux locomoteurs cervico-lombaires sur l'activité des neurones respiratoires spinaux et bulbaires". Thesis, Bordeaux 2, 2013. http://www.theses.fr/2013BOR22089/document.
Pełny tekst źródłaThe central nervous system contains neural networks that can generate rhythmic motor drive in absence of sensory feedback. These neural networks are commonly called central pattern generators (CPG) and are involved in many vital functions and behaviors, such as locomotion or respiration. In certain circumstances, these neural networks must interact to produce motor behaviors adapted to environmental constraints and the basic needs of organism. This is the case during physical exercise when the respiratory frequency increases in order to satisfy the oxygen needs. In a context of integrative neurosciences, my doctoral work aimed at exploring the neurogenic mechanisms involved in the coordination between the medullary respiratory networks and the spinal locomotor CPG. To address this question, we used an isolated in vitro brain stem-spinal cord preparations from neonatal rats (0-2 days) in which the respiratory and the locomotor networks are kept intact. Using electrophysiological, pharmacological, lesional and neuroanatomical approaches, mechanisms involved in the coordination between locomotor and respiratory rhythms have been studied. The major finding of this doctoral work is the identification of an ascending excitatory influence from spinal locomotor CPG to the respiratory networks, acting particularly on the parafacial respiratory group, which is known to be engaged in the genesis of expiratory activity. In addition to the respiratory frequency modulation, this ascending influence also modulates the activity of spinal expiratory neurons located in lumbar and thoracic segments. These data provide the first evidence for the existence of bi-functional neurons in newborn rat spinal cord
Lachaud, Guillaume. "Extensions and Applications of Graph Neural Networks". Electronic Thesis or Diss., Sorbonne université, 2023. http://www.theses.fr/2023SORUS434.
Pełny tekst źródłaGraphs are used everywhere to represent interactions between entities, whether physical such as atoms, molecules or people, or more abstract such as cities, friendships, ideas, etc. Amongst all the methods of machine learning that can be used, the recent advances in deep learning have made graph neural networks the de facto standard for graph representation learning. This thesis can be divided in two parts. First, we review the theoretical underpinnings of the most powerful graph neural networks. Second, we explore the challenges faced by the existing models when training on real world graph data. The powerfulness of a graph neural network is defined in terms of its expressiveness, i.e., its ability to distinguish non isomorphic graphs; or, in an equivalent manner, its ability to approximate permutation invariant and equivariant functions. We distinguish two broad families of the most powerful models. We summarise the mathematical properties as well as the advantages and disadvantages of these models in practical situations. Apart from the choice of the architecture, the quality of the graph data plays a crucial role in the ability to learn useful representations. Several challenges are faced by graph neural networks given the intrinsic nature of graph data. In contrast to typical machine learning methods that deal with tabular data, graph neural networks need to consider not only the features of the nodes but also the interconnectedness between them. Due to the connections between nodes, training neural networks on graphs can be done in two settings: in transductive learning, the model can have access to the test features in the training phase; in the inductive setting, the test data remains unseen. We study the differences in terms of performance between inductive and transductive learning for the node classification task. Additionally, the features that are fed to a model can be noisy or even missing. In this thesis we evaluate these challenges on real world datasets, and we propose a novel architecture to perform missing data imputation on graphs. Finally, while graphs can be the natural way to describe interactions, other types of data can benefit from being converted into graphs. In this thesis, we perform preliminary work on how to extract the most important parts of skin lesion images that could be used to create graphs and learn hidden relations in the data
Bugnicourt, Ghislain. "Adhésion, croissance et polarisation de neurones sur substrats micro-et nano-structurés". Phd thesis, Université de Grenoble, 2011. http://tel.archives-ouvertes.fr/tel-00665074.
Pełny tekst źródłaBelblidia, Nadjet. "Capacité des réseaux opportunistes : caractérisation et impact sur la dissémination de contenus volumineux". Paris 6, 2012. http://www.theses.fr/2012PA066064.
Pełny tekst źródłaA common assumption in intermittently-connected (or opportunistic) mobile networks is that any contact has enough capacity to transfer any amount of data. Although such an assumption is reasonable for analytical purposes and when contents are small, it does not hold anymore when users produce contents that are larger than the contact capacity. In the first part of this thesis, we define and evaluate the surround indicator as a metric to exhibit a contact's surrounding environment in opportunistic networks. Whereas communication opportunities are limited in terms of both duration and surrounding environment, users conversely generate, consume, and share contents that are becoming increasingly larger. In such a situation, content-sharing solutions must be reformulated to enable exchanging large contents. Users must slice data and send fragments separately, which leads to a better use of short-lived contacts and promotes progressive dissemination of large contents. The main question here is to design the best strategy for deciding which piece(s) to transmit whenever nodes meet. We present the design and evaluation of PACS, a completely distributed algorithm that selects pieces to transfer based on a local view of their popularity. Finally, we present some experimental results obtained using PePiT, an Android application based on PACS that enables the dissemination of multimedia files among collocated Android devices in ad hoc mode. Then, we go one step further in the investigation of large content dissemination challenges in opportunistic networks. We argue that uniform random inter-content selection may not be sufficient in real-world deployment. We propose EPICS, a distributed strategy that enables fulfilling dissemination policy objectives. EPICS is based on the grey relational analysis to weight content selection. In our study, we use EPICS to reduce the dissemination delay variability due to the uniform random inter-content selection
Hagenbach, Jeanne. "Communication stratégique et réseaux". Phd thesis, Université Panthéon-Sorbonne - Paris I, 2009. http://tel.archives-ouvertes.fr/tel-00450632.
Pełny tekst źródłaHachi, Ryma. "Explorer l'effet de la morphologie des réseaux viaires sur leurs conditions d'accessibilité : une approche empirique fondée sur la théorie des graphes". Thesis, Paris 1, 2020. http://www.theses.fr/2020PA01H072.
Pełny tekst źródłaThis thesis aims to explore the relationship between the morphology of street networks and the accessibility offered to individuals during their trips in the urban space. The accessibility is defined as a set of favourable conditions for traveling (e.g. short distances to cover, low congestion level). This relationship is the subject of much tacit knowledge in the urban design community. Typical network morphologies or typical interventions on existing networks are recommended by urban designers, for the accessibility conditions they are supposed to offer. However, the actual effects of these recommendations on accessibility conditions are little evaluated in a formalized and systematic way. To compensate for this lack, we choose to adopt a quantitative approach based on graph theory. This allows an analysis of the morphology and accessibility conditions of networks by means of descriptors calculated on graphs, and then the study of the relationship between morphological and accessibility descriptors. Our work is exploratory. It concerns a set of ten empirical case studies, chosen for their representativity of theoretical cases recommended in urban design. We have constituted two corpuses of study. The first brings together networks with a typical morphology. This is the case of organic networks such as Paris in the Middle Ages, grid networks like Manhattan, and tree-like networks like in some American suburbs. The second corpus is made up of successive states of a network in which typical interventions, recommended in the literature, have been carried out. In this case, it concerns the creation of star-shaped breakthroughs in the street network of Paris in the 19th century. The quantitative description of the morphological characteristics and the accessibility conditions, carried out on the two corpuses, reveals some specificities of each typical network and intervention analyzed, both in terms of morphology and accessibility. Furthermore, our results allow us to identify trends in the relationship between the morphological characteristics of the studied networks and their accessibility conditions. In particular, we show that these trends are more marked for the corpus of networks with a typical morphology than for the Parisian network at different dates : in Paris, strong variations in morphological descriptors are often accompanied by weak variations in accessibility descriptors. From a thematic point of view, this result suggests that the major works carried out in the 19th century by Haussmann certainly affected the morphology of the street network, but had a little effect on the accessibility conditions offered by this network. Eventually, we conclude that the adoption of a quantitative approach to deal with the relationship between the morphology of a street network and its accessibility conditions requires a back and forth movement between the knowledge and interpretations specific to urban design and the methods and measures from other disciplines, in this case network science
Seigneur, Josée. "Impacts des rythmes du sommeil sur la connectivité fonctionnelle et effets des changements ioniques sur la synchronisation neuronale et la connectivité fonctionnelle". Thesis, Université Laval, 2013. http://www.theses.ulaval.ca/2013/29935/29935.pdf.
Pełny tekst źródłaThe neuronal synchronisation is an intrinsic phenomenon in the brain that allows neurons to be connected to the network to communicate. Oscillations inherent of the states of vigilance such as the slow-wave sleep, the REM sleep, and waking state or pathological conditions such as epilepsy emerge from the network synchronisation of a group of neurons. Several interactions influence the synchronization: the chemical or electrical transmission, the ionic variations, and the ephaptic interactions. At cellular level, the synaptic plasticity also influences the functional connectivity of neurons. In this thesis, I aim to explain the impact of sleep rhythms on the functional connectivity and the effects of ionic variations on the neuronal synchrony and the functional connectivity. States of vigilance implicated in the memory consolidation. We demonstrated that the presence of slow oscillations and the spiking pattern during slow-wave sleep favours the long-term synaptic facilitation, which could be a key element for the sleep-dependent reinforcement of synaptic efficacy contributing to memory consolidation. By contrast synaptic activities generated during waking state in a conditions of increased level acetylcholine favour short-term facilitation. Sleep allows also the brain to disrupt partially the communication with the environment. The accepted model is that the thalamus provides gating of sensory information during sleep, but the exact mechanisms of that gating are unknown. We demonstrated that the failure rate to a lemniscal stimulation is increased during the thalamic Ca2+ spike bursts and the generation of those Ca2+ spikes cause a depletion of the extracellular calcium which is sufficient to reduce the synaptic efficacy. Bursts of action potential occur preferentially during slow-wave sleep, but also in the pathological form of paroxysmal depolarization shift during the generation of cortical epileptic seizures. During seizures, the paroxysmal neuronal activity causes a decrease of extracellular Ca2+ and an increase of extracellular potassium. We demonstrated that those ionic variations affect the synaptic transmission by increasing the failure rate of unitary responses at a synapse and by blocking the axonal transmission of action potentials, which disrupts the neuronal communication between neurons, facilitating seizure termination.
Zouhri, Abdelhakim. "Essai sur les indicateurs avancés de risque-pays : application des réseaux de neurones et choix de politiques optimales". Nice, 2007. http://www.theses.fr/2007NICE0036.
Pełny tekst źródłaThis study suggests a methodology to measure specific risks in emerging countries. Nowadays this appears even more critical as most investors adopt these risk measures in an international investments selection process while the focus is on profitability of the investisments. The idea behind this model lives on a theoretical frame formulation that provides us with a practical method to obtain and ckeck out these keys points of the probability to a case of country crisis in the main emerging countries. Norals networks gether results which have implications on the sturdiness of results presented by the crisis model of the third generation exchange in terms of efficiency of monetary politic
Lauret, Pierre. "Modélisation de la dispersion atmosphérique sur un site industriel par combinaison d’automates cellulaires et de réseaux de neurones". Thesis, Saint-Etienne, EMSE, 2014. http://www.theses.fr/2014EMSE0745/document.
Pełny tekst źródłaAtmospheric dispersion of hazardous materials is an event that could lead to serious consequences. Atmospheric dispersion is studied in particular in this work. Modeling of atmospheric dispersion is an important tool to anticipate industrial accidents. The objective of this work was to develop a model that is both fast and accurate, considering the dispersion in the near field on an industrial site. The approach developed is based on models from artificial intelligence: neural networks and cellular automata. Using neural networks requires training a database typical of the phenomenon, CFD k-ϵ simulations in this work. Training the neural network is carried out by identifying the important parameters: database sampling and network architecture. Three methodologies are developed:The first method estimates the continuous dispersion in free field by neural networks.The second method uses the neural network as a transition rule of the cellular automaton to estimate puff evolution in the free field.The third method divides the problem: the flow calculation is performed by the neural network and the calculation of the dispersion is realized by solving the advection diffusion equation to estimate the evolution of a cloud around a cylindrical obstacle. For the three methods, assessment of the generalization capabilities of the neural network has been validated on a test database and on unlearned cases. A comparison between developed method and CFD simulations is done on unlearned cases in order to validate them. Simulations computing time are low according to crisis duration. Application to real data should be developed to make these models operational
Niepceron, Brad. "Développement d'une application d'aide au diagnostic basée sur les réseaux de neurones artificiels pour la détection de tumeurs cérébrales". Electronic Thesis or Diss., Amiens, 2021. http://www.theses.fr/2021AMIE0071.
Pełny tekst źródłaDuring the last decade, the study of brain tumor diagnosis systems brought a significant attention regarding to the fast growth of deep learning and the development of Artificial Neural Networks (ANNs). In the clinical field, deep learning based algorithms are being used to solve visual tasks such as the detection and segmentation of unhealthy tissues. These methods proved to be particularly efficient in the diagnosis of aggressive tumors like high grade gliomas. However, constrained by their important need in computational resources, these models cannot be realistically deployed on a large scale.In fact, their architecture becoming deeper with the improvement of their performances, their use and development entails significant material and energy costs as well as an important carbon dioxide emission. The optimization or replacement of these methods by solutions that are less dependent on the availability of high computational resources is thus critical. To respond to these problems, the compression of modern Convolutional Neural Networks (CNNs) for the creation of brain tumor segmentation applications on embedded systems is considered. Moreover, although many debates appeared concerning the efficiency of Deep Learning algorithms, some solutions based on Spiking Neural Networks (SNNs) are yet to be investigated in order to build fast and affordable medical image analysis systems.The objective of this work is thus to propose new ways to design medical image analysis systems, specifically for glioma tumors diagnosis. We aim to tackle the computational and energy cost issues of existing deep learning solutions to let their deployment be realistic in clinical settings. Hence, the first contribution presented in this manuscript firstly focuses on the adaptation of ANNs to devices with limited computational resources by the means of compression methods. Then, in a second contribution, non-trainable neural models for medical image analysis are investigated in order to respond to the cost problems induced by deep learning. Finally, our third contribution present a new method for the development of brain tumor diagnosis systems based on models of biological neurons
Pinna, Andrea. "Conception d'une rétine connexionniste : du capteur au système de vision sur puce". Paris 6, 2003. http://www.theses.fr/2003PA066566.
Pełny tekst źródłaBornat, Yannick. "Réseaux de neurones sur silicium : une approche mixte, analogique / numérique, pour l'étude des phénomènes d'adaptation, d'apprentissage et de plasticité". Phd thesis, Université Sciences et Technologies - Bordeaux I, 2006. http://tel.archives-ouvertes.fr/tel-00181353.
Pełny tekst źródła