Dissertations / Theses on the topic 'Systèmes embarqués (informatique) – Apprentissage automatique'
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 'Systèmes embarqués (informatique) – Apprentissage automatique.'
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.
Ouairy, Léopold. "Analyse des vulnérabilités dans des systèmes embarqués face à des attaques par fuzzing." Thesis, Rennes 1, 2020. http://www.theses.fr/2020REN1S020.
Full textNowadays, smart cards are used daily. They enable users to pay or sign numeric documents for example. Because they contain sensible information about their user and secrets, attackers are interested in them. In particular, these attackers can use fuzzing. This attack consists in sending the most possible communication messages to a program in order to detect its vulnerabilities. This thesis aims at protecting smart cards against fuzzing. Two approaches for detecting implementation errors are proposed. The first one is from the state of the art, and it is adapted and improved for Java. It is based on an automated source code analysis. The second approach analyses source codes too, but it takes into account limitations of the first one. In particular, the precision and the dimension reduction is improved by using Natural Language Processing techniques. In addition, it studies plagiarsm techniques in order to reinforce its analysis against different implementations choices. An inter-procedural of the control flow graph is achieved for reducing the false positives. Both approaches are evaluated against three manually created oracles from OpenPGP and AES implementations for the neighborhood discovery and the anomaly detection. Results show that the second approach is improved in precision, recall and with less execution time than the first one. Its implementation, Confiance can be used in companies to secure source codes
Godin, Christelle. "Contributions à l'embarquabilité et à la robustesse des réseaux de neurones en environnement radiatif : apprentissage constructif : neurones à impulsions." École nationale supérieure de l'aéronautique et de l'espace (Toulouse ; 1972-2007), 2000. http://www.theses.fr/2000ESAE0013.
Full textAllenet, Thibault. "Quantization and adversarial robustness of embedded deep neural networks." Electronic Thesis or Diss., Université de Rennes (2023-....), 2023. https://ged.univ-rennes1.fr/nuxeo/site/esupversions/5f524c49-7a4a-4724-ae77-9afe383b7c3c.
Full textConvolutional Neural Networks (CNNs) and Recurrent Neural Networks (RNNs) have been broadly used in many fields such as computer vision, natural language processing and signal processing. Nevertheless, the computational workload and the heavy memory bandwidth involved in deep neural networks inference often prevents their deployment on low-power embedded devices. Moreover, deep neural networks vulnerability towards small input perturbations questions their deployment for applications involving high criticality decisions. This PhD research project objective is twofold. On the one hand, it proposes compression methods to make deep neural networks more suitable for embedded systems with low computing resources and memory requirements. On the other hand, it proposes a new strategy to make deep neural networks more robust towards attacks based on crafted inputs with the perspective to infer on edge. We begin by introducing common concepts for training neural networks, convolutional neural networks, recurrent neural networks and review the state of the art neural on deep neural networks compression methods. After this literature review we present two main contributions on compressing deep neural networks: an investigation of lottery tickets on RNNs and Disentangled Loss Quantization Aware Training (DL-QAT) on CNNs. The investigation of lottery tickets on RNNs analyze the convergence of RNNs and study its impact when subject to pruning on image classification and language modelling. Then we present a pre-processing method based on data sub-sampling that enables faster convergence of LSTM while preserving application performance. With the Disentangled Loss Quantization Aware Training (DL-QAT) method, we propose to further improve an advanced quantization method with quantization friendly loss functions to reach low bit settings like binary parameters where the application performance is the most impacted. Experiments on ImageNet-1k with DL-QAT show improvements by nearly 1\% on the top-1 accuracy of ResNet-18 with binary weights and 2-bit activations, and also show the best profile of memory footprint over accuracy when compared with other state-of-the art methods. This work then studies neural networks robustness toward adversarial attacks. After introducing the state of the art on adversarial attacks and defense mechanisms, we propose the Ensemble Hash Defense (EHD) defense mechanism. EHD enables better resilience to adversarial attacks based on gradient approximation while preserving application performance and only requiring a memory overhead at inference time. In the best configuration, our system achieves significant robustness gains compared to baseline models and a loss function-driven approach. Moreover, the principle of EHD makes it complementary to other robust optimization methods that would further enhance the robustness of the final system and compression methods. With the perspective of edge inference, the memory overhead introduced by EHD can be reduced with quantization or weight sharing. The contributions in this thesis have concerned optimization methods and a defense system to solve an important challenge, that is, how to make deep neural networks more robust towards adversarial attacks and easier to deployed on the resource limited platforms. This work further reduces the gap between state of the art deep neural networks and their execution on edge devices
Ramu, Jean-Philippe. "Efficience d'une documentation opérationnelle contextuelle sur la performance des pilotes de transport aérien." Toulouse, ISAE, 2008. http://www.theses.fr/2008ESAE0020.
Full textEmteu, Tchagou Serge Vladimir. "Réduction à la volée du volume des traces d'exécution pour l'analyse d'applications multimédia de systèmes embarqués." Thesis, Université Grenoble Alpes (ComUE), 2015. http://www.theses.fr/2015GREAM051/document.
Full textThe consumer electronics market is dominated by embedded systems due to their ever-increasing processing power and the large number of functionnalities they offer.To provide such features, architectures of embedded systems have increased in complexity: they rely on several heterogeneous processing units, and allow concurrent tasks execution.This complexity degrades the programmability of embedded system architectures and makes application execution difficult to understand on such systems.The most used approach for analyzing application execution on embedded systems consists in capturing execution traces (event sequences, such as system call invocations or context switch, generated during application execution).This approach is used in application testing, debugging or profiling.However in some use cases, execution traces generated can be very large, up to several hundreds of gigabytes.For example endurance tests, which are tests consisting in tracing execution of an application on an embedded system during long periods, from several hours to several days.Current tools and methods for analyzing execution traces are not designed to handle such amounts of data.We propose an approach for monitoring an application execution by analyzing traces on the fly in order to reduce the volume of recorded trace.Our approach is based on features of multimedia applications which contribute the most to the success of popular devices such as set-top boxes or smartphones.This approach consists in identifying automatically the suspicious periods of an application execution in order to record only the parts of traces which correspond to these periods.The proposed approach consists of two steps: a learning step which discovers regular behaviors of an application from its execution trace, and an anomaly detection step which identifies behaviors deviating from the regular ones.The many experiments, performed on synthetic and real-life datasets, show that our approach reduces the trace size by an order of magnitude while maintaining a good performance in detecting suspicious behaviors
Bahl, Gaétan. "Architectures deep learning pour l'analyse d'images satellite embarquée." Thesis, Université Côte d'Azur, 2022. https://tel.archives-ouvertes.fr/tel-03789667.
Full textThe recent advances in high-resolution Earth observation satellites and the reduction in revisit times introduced by the creation of constellations of satellites has led to the daily creation of large amounts of image data hundreds of TeraBytes per day). Simultaneously, the popularization of Deep Learning techniques allowed the development of architectures capable of extracting semantic content from images. While these algorithms usually require the use of powerful hardware, low-power AI inference accelerators have recently been developed and have the potential to be used in the next generations of satellites, thus opening the possibility of onboard analysis of satellite imagery. By extracting the information of interest from satellite images directly onboard, a substantial reduction in bandwidth, storage and memory usage can be achieved. Current and future applications, such as disaster response, precision agriculture and climate monitoring, would benefit from a lower processing latency and even real-time alerts.In this thesis, our goal is two-fold: On the one hand, we design efficient Deep Learning architectures that are able to run on low-power edge devices, such as satellites or drones, while retaining a sufficient accuracy. On the other hand, we design our algorithms while keeping in mind the importance of having a compact output that can be efficiently computed, stored, transmitted to the ground or other satellites within a constellation.First, by using depth-wise separable convolutions and convolutional recurrent neural networks, we design efficient semantic segmentation neural networks with a low number of parameters and a low memory usage. We apply these architectures to cloud and forest segmentation in satellite images. We also specifically design an architecture for cloud segmentation on the FPGA of OPS-SAT, a satellite launched by ESA in 2019, and perform onboard experiments remotely. Second, we develop an instance segmentation architecture for the regression of smooth contours based on the Fourier coefficient representation, which allows detected object shapes to be stored and transmitted efficiently. We evaluate the performance of our method on a variety of low-power computing devices. Finally, we propose a road graph extraction architecture based on a combination of fully convolutional and graph neural networks. We show that our method is significantly faster than competing methods, while retaining a good accuracy
Morozkin, Pavel. "Design and implementation of image processing and compression algorithms for a miniature embedded eye tracking system." Electronic Thesis or Diss., Sorbonne université, 2018. https://accesdistant.sorbonne-universite.fr/login?url=https://theses-intra.sorbonne-universite.fr/2018SORUS435.pdf.
Full textHuman-Machine Interaction (HMI) progressively becomes a part of coming future. Being an example of HMI, embedded eye tracking systems allow user to interact with objects placed in a known environment by using natural eye movements. The EyeDee™ portable eye tracking solution (developed by SuriCog) is an example of an HMI-based product, which includes Weetsy™ portable wire/wireless system (including Weetsy™ frame and Weetsy™ board), π-Box™ remote smart sensor and PC-based processing unit running SuriDev eye/head tracking and gaze estimation software, delivering its result in real time to a client’s application through SuriSDK (Software Development Kit). Due to wearable form factor developed eye-tracking system must conform to certain constraints, where the most important are low power consumption, low heat generation low electromagnetic radiation, low MIPS (Million Instructions per Second), as well as support wireless eye data transmission and be space efficient in general. Eye image acquisition, finding of the eye pupil ROI (Region Of Interest), compression of ROI and its wireless transmission in compressed form over a medium are very beginning steps of the entire eye tracking algorithm targeted on finding coordinates of human eye pupil. Therefore, it is necessary to reach the highest performance possible at each step in the entire chain. In contrast with state-of-the-art general-purpose image compression systems, it is possible to construct an entire new eye tracking application-specific image processing and compression methods, approaches and algorithms, design and implementation of which are the goal of this thesis
Yang, Wenlu. "Personalized physiological-based emotion recognition and implementation on hardware." Thesis, Sorbonne université, 2018. http://www.theses.fr/2018SORUS064.
Full textThis thesis investigates physiological-based emotion recognition in a digital game context and the feasibility of implementing the model on an embedded system. The following chanllenges are addressed: the relationship between emotional states and physiological responses in the game context, individual variabilities of the pschophysiological responses and issues of implementation on an embedded system. The major contributions of this thesis are : Firstly, we construct a multi-modal Database for Affective Gaming (DAG). This database contains multiple measurements concerning objective modalities: physiological signals (ECG, EDA, EMG, Respiration), screen recording, and player's face recording, as well as subjective assessments on both game event and match level. We presented statistics of the database and run a series of analysis on issues such as emotional moment detection and emotion classification, influencing factors of the overall game experience using various machine learning methods. Secondly, we investigate the individual variability in the collected data by creating an user-specific model and analyzing the optimal feature set for each individual. We proposed a personalized group-based model created the similar user groups by using the clustering techniques based on physiological traits deduced from optimal feature set. We showed that the proposed personalized group-based model performs better than the general model and user-specific model. Thirdly, we implemente the proposed method on an ARM A9 system and showed that the proposed method can meet the requirement of computation time
Caron, François. "Conception automatique de systèmes embarqués pour la téléphonie mobile 3G." Nice, 2006. http://www.theses.fr/2006NICE4091.
Full textDue to new request from the market of the mobile phone, the new third generation of standards challenges designers. Actually, these standards define a multitude of services using different bitrates that the final customer can use. This capability implies a very high complexity at some points of the network, in particular for base-stations. Base stations of the third generation mobile networks should be able to process a lot of data received from and sending to a multitude of users, communicating at the same time. PhD work presented in this document is about the definition of a new software/hardware co-design tool. This tool called BERLIOZ, « emBedded systEM exploRation for pipeLIned executiOn optimiZation », offers to embedded systems designers a set of architectural solutions to the problems posed by third mobile generations standards like the UMTS. So, this tool can analyse applications of signal processing which periodicity can be shorter than their deadlines. The computation of solutions is done with the help of a heuristic composed of two genetic algorithms. The use of these algorithms allows to achieve solutions matching a set of constraints like the silicium area of a system. After the deep study of third generation standard called UMTS, we have validated our approach on a real industrial application : the design of the “Symbol rate” part of a UMTS base-station
Lévy, Christophe. "Modèles acoustiques compacts pour les systèmes embarqués." Avignon, 2006. http://www.theses.fr/2006AVIG0143.
Full textThe amount of services offered by the last generation mobile phones has significantly increased compared to previous generations. Nowadays, phones offer new kinds of facilitiessuch as organizers, phone books, e-mail/fax, and games. At the same time, the size of mobile phones has steadily reduced. Both these observations raise an important question: ?How can we use the full facilities of a mobile phone without a large keyboard??. Voice based human-to-computer interfaces supply a friendly solution to this problem but require an embedded speech recognizer. Over the last decade, the performance of Automatic Speech Recognition (ASR) systems has improved and nowadays facilites the implementation of vocal human-to-computer interfaces. Moreover, even if scientific progress could be noticed, the potential gain (in performance) remains limited by computing resources: a relatively modern computer with a lot of memory is generally required. The main problem to embed ASR in a mobile phone is the low level of resources available in this context which classically consists of a 50/100 MHz processor, a 50/100 MHz DSP, and less than 100KB of memory. This thesis focuses on embedded speech recognition in the context of limited resources
Almousa, Almaksour Abdullah. "Apprentissage incrémental de systèmes d'inférence floue : Application à la reconnaissance de gestes manuscrits." Rennes, INSA, 2011. http://www.theses.fr/2011ISAR0015.
Full textWe present in a new method for the conception of evolving and customizable classification systems. We propose an incremental approach for the learning of classification models based on first-order Takagi-Sugeno (TS) fuzzy inference systems. This approach includes, on the one hand, the adaptation of linear consequences of the fuzzy rules using the recursive least-squares method, and, on the other hand, an incremental learning of the antecedent of these rules in order to modify the membership functions according to the evolution of data density in the input space. The proposed method resolves the instability problems in the incremental learning of TS models thanks to a global learning paradigm in which antecedent and consequents are learned in synergy, contrary to the existing approaches where they are learned separately. The performance of our system had been demonstrated on different well-known benchmarks, with a special focus on its capacity of learning new classes
Pradel, Bruno. "Evaluation des systèmes de recommandation à partir d'historiques de données." Paris 6, 2013. http://www.theses.fr/2013PA066263.
Full textThis thesis presents various experimental protocols leading to abetter offline estimation of errors in recommender systems. As a first contribution, results form a case study of a recommendersystem based on purchased data will be presented. Recommending itemsis a complex task that has been mainly studied considering solelyratings data. In this study, we put the stress on predicting thepurchase a customer will make rather than the rating he will assign toan item. While ratings data are not available for many industries andpurchases data widely used, very few studies considered purchasesdata. In that setting, we compare the performances of variouscollaborative filtering models from the litterature. We notably showthat some changes the training and testing phases, and theintroduction of contextual information lead to major changes of therelative perfomances of algorithms. The following contributions will focus on the study of ratings data. Asecond contribution will present our participation to the Challenge onContext-Aware Movie Recommendation. This challenge provides two majorchanges in the standard ratings prediction protocol: models areevaluated conisdering ratings metrics and tested on two specificsperiod of the year: Christmas and Oscars. We provides personnalizedrecommendation modeling the short-term evolution of the popularitiesof movies. Finally, we study the impact of the observation process of ratings onranking evaluation metrics. Users choose the items they want to rateand, as a result, ratings on items are not observed at random. First,some items receive a lot more ratings than others and secondly, highratings are more likely to be oberved than poor ones because usersmainly rate the items they likes. We propose a formal analysis ofthese effects on evaluation metrics and experiments on the Yahoo!Musicdataset, gathering standard and randomly collected ratings. We showthat considering missing ratings as negative during training phaseleads to good performances on the TopK task, but these performancescan be misleading favoring methods modeling the popularities of itemsmore than the real tastes of users
Chkouri, Mohamed Yassin. "Modélisation des systèmes temps-réel embarqués en utilisant AADL pour la génération automatique d'applications formellement vérifiées." Phd thesis, Grenoble, 2010. http://tel.archives-ouvertes.fr/tel-00516152.
Full textVergnaud, Thomas. "Modélisation des systèmes temps-réel répartis embarqués pour la génération automatique d'applications formellement vérifiées." Phd thesis, Télécom ParisTech, 2006. http://pastel.archives-ouvertes.fr/pastel-00002122.
Full textDomingues, Antonio. "Système embarqué multi-capteurs pour la détection d'obstacles routiers : développement du prototype et réglage automatique de la chaîne de traitements d'images." Orléans, 2004. http://www.theses.fr/2004ORLE2026.
Full textBoucheron, Stéphane. "Apprentissage et calculs." Montpellier 2, 1988. http://www.theses.fr/1988MON20251.
Full textRoynard, Xavier. "Sémantisation à la volée de nuages de points 3D acquis par systèmes embarqués." Thesis, Paris Sciences et Lettres (ComUE), 2019. http://www.theses.fr/2019PSLEM078.
Full textThis thesis is at the confluence of two worlds in rapid growth: autonomous cars and artificial intelligence (especially deep learning). As the first takes advantage of the second, autonomous vehicles are increasingly using deep learning methods to analyze the data produced by its various sensors (including LiDARs) and to make decisions. While deep learning methods have revolutionized image analysis (in classification and segmentation for example), they do not produce such spectacular results on 3D point clouds. This is particularly true because the datasets of annotated 3D point clouds are rare and of moderate quality. This thesis therefore presents a new dataset developed by mobile acquisition to produce enough data and annotated by hand to ensure a good quality of segmentation. In addition, these datasets are inherently unbalanced in number of samples per class and contain many redundant samples, so a sampling method adapted to these datasets is proposed. Another problem encountered when trying to classify a point from its neighbourhood as a voxel grid is the compromise between a fine discretization step (for accurately describing the surface adjacent to the point) and a large grid (to look for context a little further away). We therefore also propose network methods that take advantage of multi-scale neighbourhoods. These methods achieve the state of the art of point classification methods on public benchmarks. Finally, to respect the constraints imposed by embedded systems (real-time processing and low computing power), we present a method that allows convolutional layers to be applied only where there is information to be processed
Hajjar, Salam. "Conception sûre de systèmes embarqués à base de COTS." Phd thesis, INSA de Lyon, 2013. http://tel.archives-ouvertes.fr/tel-00952827.
Full textAbdelmoula, Mariem. "Génération automatique de jeux de tests avec analyse symbolique des données pour les systèmes embarqués." Thesis, Nice, 2014. http://www.theses.fr/2014NICE4149/document.
Full textOne of the biggest challenges in hardware and software design is to ensure that a system is error-free. Small errors in reactive embedded systems can have disastrous and costly consequences for a project. Preventing such errors by identifying the most probable cases of erratic system behavior is quite challenging. Indeed, tests in industry are overall non-exhaustive, while formal verification in scientific research often suffers from combinatorial explosion problem. We present in this context a new approach for generating exhaustive test sets that combines the underlying principles of the industrial test technique and the academic-based formal verification approach. Our approach builds a generic model of the system under test according to the synchronous approach. The goal is to identify the optimal preconditions for restricting the state space of the model such that test generation can take place on significant subspaces only. So, all the possible test sets are generated from the extracted subspace preconditions. Our approach exhibits a simpler and efficient quasi-flattening algorithm compared with existing techniques and a useful compiled internal description to check security properties and reduce the state space combinatorial explosion problem. It also provides a symbolic processing technique of numeric data that provides a more expressive and concrete test of the system. We have implemented our approach on a tool called GAJE. To illustrate our work, this tool was applied to verify an industrial project on contactless smart cards security
Nagoudi-Layerle, Khaoula. "Application de la commande prédictive reconfigurable sur le circuit d'air du moteur Diesel." Rouen, 2011. http://www.theses.fr/2011ROUES049.
Full textLe, Roux Bernard. "Eléments d'une approche constructive de la modélisation et de la réutilisation en acquisition des connaissances." Paris 6, 1994. http://www.theses.fr/1994PA066628.
Full textMartinez, Dominique. ""Offset", une méthode de construction incrémentale de réseaux de neurones multicouches et son application à la conception d'un copilote automobile." Toulouse 3, 1992. http://www.theses.fr/1992TOU30033.
Full textLurette, Christophe. "Développement d'une technique neuronale auto-adaptative pour la classification dynamique de données évolutives : application à la supervision d'une presse hydraulique." Lille 1, 2003. https://ori-nuxeo.univ-lille1.fr/nuxeo/site/esupversions/aed48e38-323f-425b-b6ff-c8e75ff5d4b6.
Full textDamas, Luc. "Étude théorique et pratique de la production d'effets d'amorçage de la mémoire : application à l'assistance à la remémoration chez l'utilisateur d'un système informatique pour une tâche d'apprentissage." Lyon 1, 2003. http://www.theses.fr/2003LYO10226.
Full textSarzeaud, Olivier. "Contribution à l'étude et au développement de systèmes connexionnistes modulaires." Montpellier 2, 1993. http://www.theses.fr/1993MON20137.
Full textBen, Abdelali Abdessalem. "Etude de la conception d’architectures matérielles dédiées pour les traitements multimédia : indexation de la vidéo par le contenu." Dijon, 2007. http://www.theses.fr/2007DIJOS075.
Full textThis thesis constitutes a contribution to the study of content based automatic video indexing aiming at designing hardware architectures dedicated to this type of multimedia application. The content based video indexing represents an important domain that is in constant development for different types of applications such as the Internet, the interactive TV, the personal video recorders (PVR) and the security applications. The proposed study is done through concrete AV analysis techniques for video indexing and it is carried out according to different aspects related to application, technology and methodology. It is included in the context of dedicated hardware architectures design and exploitation of the new embedded systems technologies for the recent multimedia applications. Much more interest is given to the reconfigurable technology and to the new possibilities and means of the FPGA devices utilization. The first stage of this thesis is devoted to the study of the automatic content based video indexing domain. It is about the study of features and the new needs of indexing systems through the approaches and techniques currently used as well as the application fields of the new generations of these systems. This is in order to show the interest of using new architectures and technological solutions permitting to support the new requirements of this domain. The second stage is dedicated to the validation and the optimization of some visual descriptors of the MPEG-7 standard for the video temporal segmentation. This constitutes a case study through an important example of AV content analysis techniques. The proposed study constitutes also a stage of preparation for the hardware implementation of these techniques in the context of hardware accelerators design for real time automatic video indexing. Different Algorithm Architecture Adequacy aspects have been studied through the proposition of various algorithmic transformations that can be applied for the considered algorithms. The third stage of this thesis is devoted to study the design of dedicated hardware operators for video content analysis techniques as well as the exploitation of the new reconfigurable systems technologies for designing SORC dedicated to the automatic video indexing. Several hardware architectures have been proposed for the MPEG-7 descriptors and different concepts related to the exploitation of reconfigurable technology and SORC have been studied as well (methodologies and tools for designing such systems on chip, technology and methods for the dynamic and partial reconfiguration, FPGA based hardware platforms, SORC structure for video indexing, etc. )
Monsifrot, Antoine. "Utilisation du raisonnement à partir de cas et de l'apprentissage pour l'optimisation de code." Rennes 1, 2002. http://www.theses.fr/2002REN10107.
Full textDzogang, Fabon. "Représentation et apprentissage à partir de textes pour des informations émotionnelles et pour des informations dynamiques." Paris 6, 2013. http://www.theses.fr/2013PA066253.
Full textAutomatic knowledge extraction from texts consists in mapping lowlevel information, as carried by the words and phrases extracted fromdocuments, to higher level information. The choice of datarepresentation for describing documents is, thus, essential and thedefinition of a learning algorithm is subject to theirspecifics. This thesis addresses these two issues in the context ofemotional information on the one hand and dynamic information on theother. In the first part, we consider the task of emotion extraction forwhich the semantic gap is wider than it is with more traditionalthematic information. Therefore, we propose to study representationsaimed at modeling the many nuances of natural language used fordescribing emotional, hence subjective, information. Furthermore, wepropose to study the integration of semantic knowledge which provides,from a characterization perspective, support for extracting theemotional content of documents and, from a prediction perspective,assistance to the learning algorithm. In the second part, we study information dynamics: any corpus ofdocuments published over the Internet can be associated to sources inperpetual activity which exchange information in a continuousmovement. We explore three main lines of work: automaticallyidentified sources; the communities they form in a dynamic and verysparse description space; and the noteworthy themes they develop. Foreach we propose original extraction methods which we apply to a corpusof real data we have collected from information streams over the Internet
Beaugé, Lionel. "Définition de mécanismes de mémorisation pour les systèmes neuromimétiques (inspirations pluridisciplinaires)." Nancy 1, 1995. http://www.theses.fr/1995NAN10407.
Full textAl, Hajj Mohamad Rami. "Reconnaissance hors ligne de mots manuscrits cursifs par l'utilisation de systèmes hybrides et de techniques d'apprentissage automatique." Paris, ENST, 2007. http://www.theses.fr/2007ENST0020.
Full textThe automatic offline recognition of handwritten words improves human-machine interaction. It is already used in many business office applications dealing with the automatic processing of documents such as automatic post sorting, and the verification and recognition of bank check amounts. The off line recognition of cursive handwritten words remains an open problem due to difficulties such as :handwriting normalization, word segmentation into compound components and the modeling of these components. The main objective of this thesis, is to propose, design, and implement a system for the automatic offline recognition of Arabic handwritten words. The proposed approach is analytical without explicit segmentation of words into compound characters, and it is based on the stochastic HMM approach (Hidden Markov models). The method is composed of two stages : a recognition stage based on different features, and a combination stage of three HMM-based classifiers. Each individual HMM classifier uses a sliding window with a specific inclination. Different combining strategies are tested, among them the Sum rule, the Majority Vote rule and the Borda Count rule. The best combination strategy consists of using a neural network-based combining classifier. The combination of these classifiers can better cope with the writing inclination, the erroneous positions of diacritical marks and points, and the overlapping of consecutive characters in handwritten words. The reference system based on the proposed method has shown best performance at the competition organized at ICDAR 2005, where a set of state-of art systems were compared and tested on the IFN/ENIT benchmark database
Li, Huihua. "Généralisation de l'ordre et des paramètres de macro-actions par apprentissage basé sur l'explication. Extension de l'apprentissage par explications sur l'ordre partiel." Paris 6, 1992. http://www.theses.fr/1992PA066233.
Full textLasalle, Jonathan. "Génération automatique de tests à partir de modèles SysML pour la validation fonctionnelle de systèmes embarqués." Phd thesis, Université de Franche-Comté, 2012. http://tel.archives-ouvertes.fr/tel-00762053.
Full textSteux, Bruno. "RT Maps, un environnement logiciel dédié à la conception d'applications embarquées temps-réel : utilisation pour la détection automatique de véhicules par fusion radar / vision." Paris, ENMP, 2001. http://www.theses.fr/2001ENMP1013.
Full textBenouhiba, Toufik. "Approche génétique et floue pour les systèmes d'agents adaptatifs : application à la reconnaissance des scenarii." Troyes, 2005. http://www.theses.fr/2005TROY0014.
Full textThe objective of this thesis is to use minimal a priori knowledge in order to generate uncertain rules which manipulate imprecise data. The proposed architecture has been tested on a multi-agent system to recognize scenarios. The realized works are distributed into three axis: - The first one concerns uncertain reasoning with imprecise data. The evidence theory and intuitionistic fuzzy logic have been used to model such reasoning. – The second axis corresponds to classifier systems and genetic programming. The proposed approach use the power of genetic programming and combine it to classifier systems. A new learning mechanism based on evidence theory is introduced in order to use this theory as a support of reasoning. – The third axis concerns cooperation in adaptive multi-agents systems. Classifier systems have been improved by using an explicit cooperation between a number of classifier agents. We also propose a new data fusion operator based on evidence theory and adapted to the manipulated data. The developed system has been used to recognize car’s maneuvers. In fact, we have proposed a multi-agent architecture to make recognition. Maneuvers are decomposed into several layers in order to recognize them with a given granularity level
Olaru, Andrei. "Un système multi-agents sensible au contexte pour les environnements d'intelligence ambiante." Paris 6, 2011. http://www.theses.fr/2011PA066639.
Full textMengelle, Thierry. "Etude d'une architecture d'environnements d'apprentissage basés sur le concept de préceptorat avisé." Toulouse 3, 1995. http://www.theses.fr/1995TOU30099.
Full textSenoune, Redouane. "Acquisition de connaissances et apprentissage automatique : contribution pour le développement incrémental d'un système à base de connaissances pour les situations de crise : application au domaine de l'eau." Lyon, INSA, 1995. http://tel.archives-ouvertes.fr/docs/00/82/23/35/PDF/1995_Senoune_Redouane.pdf.
Full text[The subject of this thesis is the development of a Knowledge-Based System for situations of crisis. Two main research issues have been studied during the of the system : knowledge acquisition and knowledge validation. The knowledge acquisition part integrates both knowledge acquisition and machine learning techniques. As a first step, the knowledge acquisition methods have been used to identify the descriptive and strategic domain knowledge and to construct the description language to use for defining the examples needed for the machine learning. The second step is to use a machine learning technique to incrementally construct a knowledge graph using cases on interventions in situations of crisis obtained from the domain experts. Two different procedures are proposed for the exploitation phase of the system. The first procedure is the interactive use of the knowledge graph, while the second is the deductive use of the knowledge graph. The knowledge validation approach proposed is based on the interactive use of the knowledge graph and on a follow-up on expert interventions in situations of crisis. ]
Maslekar, Nitin. "Adaptive Traffic Signal Control System Based on Inter-Vehicular Communication." Rouen, 2011. http://www.theses.fr/2011ROUES046.
Full textRoques, Martine. "Apprentissage et reconnaissance de structures syntaxiques par une approche connexionniste." Paris 11, 1993. http://www.theses.fr/1993PA112433.
Full textBouhamed, Heni. "L'Apprentissage automatique : de la sélection de variables à l'apprentissage de structure d'un classifieur bayésien." Rouen, 2013. http://www.theses.fr/2013ROUES037.
Full textThe work developed in the framework of this thesis deals with the problem of processing large amounts of data in machine learning model from an examples’ database. Thus, the model constructed will serve as a tool for classifying new cases. We will particularly focus firstly, to the concept of variable selection by presenting its major strategies and propelling their shortcomings, in fact, a new filter method will be developed in this work in the aim to remedy to the identified shortcomings. Secondly, we will study the super exponential increase problem of the computational complexity of learning Bayesian classifier structure in the case of using general algorithms with no special restrictions. Indeed, referring to the formula of Robinson (Robinson, 1977), it is certain that the number of the directed acyclic graph (DAG) increases with a super exponential manner according to the increase of variables numbers. So, it is proposed in this work to develop a new approach in the aim to reduce the number of possible DAG in learning structure, without losing information. Obviously, reducing the number of DAG as possible will reduce the computational complexity of the process and therefore reducing the execution time, which will allow us to model grater information systems with the same quality of exploitation
Hren, Jean-François. "Planification optimiste pour systèmes déterministes." Thesis, Lille 1, 2012. http://www.theses.fr/2012LIL10188/document.
Full textIn the field of reinforcement learning, planning in the case of deterministic systems consists of doing a forward search using a generative model of the system so as to find the action to apply in its current state. In our case, the forward search leads us to build a look-ahead tree, its root being the current state of the system. If the computational resources are limited and unknown, we have to use an algorithm which tries to minimize its regret. In other words, an algorithm returning an action to apply which is as close as possible to the optimal one in term of quality and with respect to the computational resources used. We present the optimistic planing algorithm in the case of a discrete action space. We prove a lower and upper bound in the worst case and in a particular class of problems. Also we present two algorithms using the optimistic approach but in the case of a continuous action space
Bernauer, Éric. "Les réseaux de neurones et l'aide au diagnostic : un modèle de neurones bouclés pour l'apprentissage de séquences temporelles." Toulouse 3, 1996. http://www.theses.fr/1996TOU30277.
Full textFiandino, Maxime. "Exploration d'architectures basée sur la génération automatique de plates-formes matérielles et le portage rapide du logiciel." Grenoble INPG, 2007. http://www.theses.fr/2007INPG0053.
Full textThe proposed approach is an iterative flow in three steps. The first one is the fast development and modification of the architecture executable model. The second one is the adaptation of the embedded software. The third one is the hardware and software architecture exploration. A tool has been developed in order to create and modify quickly a hardware architecture model. It uses flexible sub-systems. One method in order to adapt the embedded software is exposed, it includes: to manually add some parameterization in the software, an automatic extraction of the architecture characteristics, the generation of the low level code sources. To finish a method allow to simulate processors at different level of simulation with their embedded software, high level for fast simulation, low level for performance measurements. Following results, hardware and software are modified and the flow can restart. This flow was tested on a real application, a parallelized H264 encoder
Koch, Guillaume. "Catégorisation automatique de documents manuscrits : Application aux courriers entrants." Rouen, 2006. http://www.theses.fr/2006ROUES033.
Full textThe work described in this document deals with the design, the development and the evaluation of a system of categorization of unconstrained handwritten documents such as handwritten incoming mails. We demonstrate that a system developed for electronic document categorization can be adapted for handwritten document categorization
Es, soufi Widad. "Modélisation et fouille des processus en vue d'assister la prise de décisions dans le contexte de la conception et la supervision des systèmes." Thesis, Paris, ENSAM, 2018. http://www.theses.fr/2018ENAM0067/document.
Full textData sets are growing rapidly because of two things. First, the fourth industrial revolution that aims to transform factories into smart entities in which cyber physical systems monitor the physical processes of the factory. Second, the need to innovate in order to achieve and maintain competitiveness. Due to this huge volume of data (Big Data), (i) design and supervision processes are becoming chaotic, (ii) data within organizations is increasingly becoming difficult to exploit and (iii) engineers are increasingly lost when making decisions. Indeed, several issues are identified in industry: (i) when researching, visualizing and exchanging information, (ii) when making decisions and (iii) when managing contextual changes. Through this research work, we propose an Intelligent and modular Decision Support System (IDSS), where each of the four modules solves one of the identified issues. Process modelling and traceability modules aim to model processes and capture how they are actualy executed. The decision support module proposes the process patterns that best fit the decision context, as well as their most significant activity parameters. The contextual change management module continuously updates the decision-making module, in order to handle the dynamic aspect of the decision context. The proposed system is fully verified and half-validated in the context of the Gontrand project, aiming at intelligent and real-time supervision of gas networks favoring the injection of green gas. In order to be fully validated, the performance of the system must be analyzed after integrating and exploitating it in a real industrial environment
Claveau, Vincent. "Acquisition automatique de lexiques sémantiques pour la recherche d'information." Phd thesis, Université Rennes 1, 2003. http://tel.archives-ouvertes.fr/tel-00524646.
Full textSobral, Rui. "Techniques et systèmes d'acquisition des connaissances." Compiègne, 1989. http://www.theses.fr/1989COMPD168.
Full textFerneda, Edilson. "Conception d'un agent rationnel et examen de son raisonnement en géométrie." Montpellier 2, 1992. http://www.theses.fr/1992MON20300.
Full textHardy, Corentin. "Contribution au développement de l’apprentissage profond dans les systèmes distribués." Thesis, Rennes 1, 2019. http://www.theses.fr/2019REN1S020/document.
Full textDeep learning enables the development of a growing number of services. However, it requires large training databases and a lot of computing power. In order to reduce the costs of this deep learning, we propose a distributed computing setup to enable collaborative learning. Future users can participate with their devices and their data without moving private data in datacenters. We propose methods to train deep neural network in this distibuted system context
Rioual, Yohann. "RL-based Energy Management for Autonomous Cyber Physical Systems." Thesis, Lorient, 2019. http://www.theses.fr/2019LORIS533.
Full textThe energy management of a cyber physical system is a difficult task because of the complexity of hardware architectures and the use of OS. In addition, these systems are deployed in changing environments where their energy harvesting capacity varies. Over time, their energy consumption is modified due to the ageing of the components. Consumption models designed in the laboratory cannot take into account all possible deployment scenarios and system aging. One approach that is developing is the use of reinforcement learning in which we no longer know the system's consumption model; but thanks to this approach, the system is still able to adapt its operation during its deployment according to the evolution of its environment. Several approaches exist in reinforcement learning. The first part of this thesis is devoted to proposing guidelines to help for selecting the most appropriate approach for a given application and target. The second part of this thesis focuses on the design of the reward we give to our system that will influence its behaviour in its environment. Two reward functions able to adjust the system’s performance according to the energy available are presented. The third and last part of this thesis explores the use of several agents to independently control the different modules of our system. This approach allows a more precise control of energy consumption, reducing memory usage compared to a single agent approach