Dissertations / Theses on the topic 'Système en 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 'Système en 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.
Burg, Bernard. "Apprentissage de règles de comportement destinées au contrôle d'un système." Paris 11, 1988. http://www.theses.fr/1988PA112375.
Full textProcess control systems have to face applications which are always more ambitions and difficult to master. In some cases it is not easy to use conventional process control techniques. With the introduction of declarative methods it is possible to start in a pragmatic way and to set an implicit formulation of the problem when no explicit formulation is available. New mechanisms can be envisioned, and we conceived a rule based controller, then the difficulty remains on the design of the rule sets. To overcome this problem, we had to use jointly some learning techniques, such as data analysis to cope with noisy data and to project them into reduced space representations. Then structural techniques allow to modelise the temporal evolution of the process control and the hidden structures. Finally, artificial intelligence machine learning techniques discover the concepts and generalise the acquired knowledge. The whole technique set is supervised by artificial intelligence, it analyses the results issued from each learning step and planes the next action to perform. Three learning strategies are used: the first one starts from the data and uses inductive learning, it proves some completeness. The second one begins with a fuzzy model and acquires rules by deduction, it brings coherency via expert knowledge. Finally the behavior rules are used and refined by means of interaction with the environment. The learning program CANDIDE performed two case studies - the speed control of a DC motor the automatic driving of a car
Robineau, Pierre. "Vers un système d'apprentissage symbolique flexible et compréhensible pour une aide à la découverte de connaissances." Avignon, 1995. http://www.theses.fr/1995AVIG0111.
Full textLin, Shiuan-Sung. "Optimisation du graphe de décodage d'un système de reconnaissance vocale par apprentissage discriminant." Paris, ENST, 2007. http://www.theses.fr/2007ENST0006.
Full textThe three main knowledge sources used in the automatic speech recognition (ASR), namely the acoustic models, a dictionary and a language model, are usually designed and optimized in isolation. Our previous work proposed a methodology for jointly tuning these parameters, based on the integration of the resources as a finite-state graph, whose transition weights are trained discriminatively. In this training framework, parameter optimization is performed on a static decoding graph, whose transition weights are iteratively adjusted. We extend our previous work to a much more complex large-vocabulary task: French radio broadcast news database (ESTER). We also propose several fast decoding techniques to make the training practical. Experiments show that a reduction of 1% absolute of word error rate (WER) can be obtained, demonstrating the effectiveness of this training framework. In addition, we also investigate the strengths and shortcomings of this approach and discuss the new directions it opens
Hoet, Shirley. "Apprentissage de la communication dans un système multi-agents ouvert, asynchrone et faiblement couplé." Paris 6, 2012. http://www.theses.fr/2012PA066511.
Full textIn a Multi-Agent System (MAS) , direct communication allows agents to exchange information, delegate tasks or negotiate by sending structured messages. In current approaches, it is generally assumed that agents know the content and recipients of the messages it has to send, and the moment in time when it should send it. However, in open and loosely coupled MAS, this hypothesis is no longer valid : agents to not "know" each other and cannot determine in advance what message to send, when and to whom. The goal of this PhD thesis is to define mechanisms for agents to learn how to communicate with other agents, based on their own goals and changes perceived in the system. First, we present an exploration algorithm coupled with a multi-agent protocol that allows agents to build the content of their messages. Second, we present a reinforcement learning mechanism that allows an agent to decide when it must communicate and what message it must send. Our algorithm is based on using a memory in which the agent can store its beliefs and the communication acts it has used. The evaluation ofour learning algorithm showed problems that come from using a general-purpose memory structure. This impacts our learning mechanism by creating a too large set of states for the algorithm to workThat is why we present a new model of memory for communication learning based on storing dates and message answers. Last, we propose a mechanism that allows the agent to build a model of communication acts, i. E. The preconditions it must satisfy to send the message and a description of the expected effects of this message on the system
Ferreira, Emmanuel. "Apprentissage automatique en ligne pour un dialogue homme-machine situé." Thesis, Avignon, 2015. http://www.theses.fr/2015AVIG0206/document.
Full textA dialogue system should give the machine the ability to interactnaturally and efficiently with humans. In this thesis, we focus on theissue of the development of stochastic dialogue systems. Thus, we especiallyconsider the Partially Observable Markov Decision Process (POMDP)framework which yields state-of-the-art performance on goal-oriented dialoguemanagement tasks. This model enables the system to cope with thecommunication ambiguities due to noisy channel and also to optimize itsdialogue management strategy directly from data with Reinforcement Learning (RL)methods.Considering statistical approaches often requires the availability of alarge amount of training data to reach good performance. However, corpora of interest are seldom readily available and collectingsuch data is both time consuming and expensive. For instance, it mayrequire a working prototype to initiate preliminary experiments with thesupport of expert users or to consider other alternatives such as usersimulation techniques.Very few studies to date have considered learning a dialogue strategyfrom scratch by interacting with real users, yet this solution is ofgreat interest. Indeed, considering the learning process as part of thelife cycle of a system offers a principle framework to dynamically adaptthe system to new conditions in an online and seamless fashion.In this thesis, we endeavour to provide solutions to make possible thisdialogue system cold start (nearly from scratch) but also to improve its ability to adapt to new conditions in operation (domain extension, new user profile, etc.).First, we investigate the conditions under which initial expertknowledge (such as expert rules) can be used to accelerate the policyoptimization of a learning agent. Similarly, we study how polarized userappraisals gathered throughout the course of the interaction can beintegrated into a reinforcement learning-based dialogue manager. Morespecifically, we discuss how this information can be cast intosocially-inspired rewards to speed up the policy optimisation for bothefficient task completion and user adaptation in an online learning setting.The results obtained on a reference task demonstrate that a(quasi-)optimal policy can be learnt in just a few hundred dialogues,but also that the considered additional information is able tosignificantly accelerate the learning as well as improving the noise tolerance.Second, we focus on reducing the development cost of the spoken language understanding module. For this, we exploit recent word embedding models(projection of words in a continuous vector space representing syntacticand semantic properties) to generalize from a limited initial knowledgeabout the dialogue task to enable the machine to instantly understandthe user utterances. We also propose to dynamically enrich thisknowledge with both active learning techniques and state-of-the-artstatistical methods. Our experimental results show that state-of-the-artperformance can be obtained with a very limited amount of in-domain andin-context data. We also show that we are able to refine the proposedmodel by exploiting user returns about the system outputs as well as tooptimize our adaptive learning with an adversarial bandit algorithm tosuccessfully balance the trade-off between user effort and moduleperformance.Finally, we study how the physical embodiment of a dialogue system in a humanoid robot can help the interaction in a dedicated Human-Robotapplication where dialogue system learning and testing are carried outwith real users. Indeed, in this thesis we propose an extension of thepreviously considered decision-making techniques to be able to take intoaccount the robot's awareness of the users' belief (perspective taking)in a RL-based situated dialogue management optimisation procedure
Ramdani, Mohammed. "Système d'induction formelle à base de connaissances imprécises." Paris 6, 1994. http://www.theses.fr/1994PA066237.
Full textRafflin, catherine. "Conception d'un système de programmation et de commande de robots mobiles par apprentissage." Montpellier 2, 1995. http://www.theses.fr/1995MON20093.
Full textGarlet, Milani Luís Felipe. "Autotuning assisté par apprentissage automatique de tâches OpenMP." Thesis, Université Grenoble Alpes, 2020. http://www.theses.fr/2020GRALM022.
Full textModern computer architectures are highly complex, requiring great programming effort to obtain all the performance the hardware is capable of delivering. Indeed, while developers know potential optimizations, the only feasible way to tell which of them is faster for some platform is to test it. Furthermore, the many differences between two computer platforms, in the number of cores, cache sizes, interconnect, processor and memory frequencies, etc, makes it very challenging to have the same code perform well over several systems. To extract the most performance, it is often necessary to fine-tune the code for each system. Consequently, developers adopt autotuning to achieve some degree of portable performance. This way, the potential optimizations can be specified once, and, after testing each possibility on a platform, obtain a high-performance version of the code for that particular platform. However, this technique requires tuning each application for each platform it targets. This is not only time consuming but the autotuning and the real execution of the application differ. Differences in the data may trigger different behaviour, or there may be different interactions between the threads in the autotuning and the actual execution. This can lead to suboptimal decisions if the autotuner chooses a version that is optimal for the training but not for the real execution of the application. We propose the use of autotuning for selecting versions of the code relevant for a range of platforms and, during the execution of the application, the runtime system identifies the best version to use using one of three policies we propose: Mean, Upper Confidence Bound, and Gradient Bandit. This way, training effort is decreased and it enables the use of the same set of versions with different platforms without sacrificing performance. We conclude that the proposed policies can identify the version to use without incurring substantial performance losses. Furthermore, when the user does not know enough details of the application to configure optimally the explore-then-commit policy usedy by other runtime systems, the more adaptable UCB policy can be used in its place
Lévy, Benjamin. "Principes et architectures pour un système interactif et agnostique dédié à l’improvisation musicale." Paris 6, 2013. http://www.theses.fr/2013PA066652.
Full textThe work presented in this thesis focuses on the conception and realization of a software capable of pertinent interaction with acoustic musicians in a collective free improvisation, that is an improvisation without any predetermined knowledge of structures, rules or style. It is extended at the end of our work with considerations on emerging properties such as pulse or a broad notion of harmony. The OMax project proposes to approach this problem of non-idiomatic improvisation by learning and mimicking the style of a musician with an agnostic and incremental knowledge model. We take this computer system as our work basis and examine carefully three aspects: the conceptual principles of the system, the software architectures for effective implementations and the real-life usage of this system in numerous testing and concerts situations. Besides a thorough study of all the conceptual elements of the system based on anthropomorphic decomposition of its parts, our main contribution is the design and realization of several variations of the OMax system. Our work has been also strongly coupled with the testing of our prototypes with several leading musicians
Nicolas, Jacques. "Ally, un systeme logique pour la generalisation en apprentissage automatique." Rennes 1, 1987. http://www.theses.fr/1987REN10043.
Full textNicolas, Jacques. "ALLY, un système logique pour la généralisation en apprentissage automatique." Grenoble 2 : ANRT, 1987. http://catalogue.bnf.fr/ark:/12148/cb37608434q.
Full textLe, Lann Marie-Véronique. "Commande prédictive et commande par apprentissage : étude d'une unité pilote d'extraction, optimisation par apprentissage." Toulouse, INPT, 1988. http://www.theses.fr/1988INPT023G.
Full textMargeta, Ján. "Apprentissage automatique pour simplifier l’utilisation de banques d’images cardiaques." Thesis, Paris, ENMP, 2015. http://www.theses.fr/2015ENMP0055/document.
Full textThe recent growth of data in cardiac databases has been phenomenal. Cleveruse of these databases could help find supporting evidence for better diagnosis and treatment planning. In addition to the challenges inherent to the large quantity of data, the databases are difficult to use in their current state. Data coming from multiple sources are often unstructured, the image content is variable and the metadata are not standardised. The objective of this thesis is therefore to simplify the use of large databases for cardiology specialists withautomated image processing, analysis and interpretation tools. The proposed tools are largely based on supervised machine learning techniques, i.e. algorithms which can learn from large quantities of cardiac images with groundtruth annotations and which automatically find the best representations. First, the inconsistent metadata are cleaned, interpretation and visualisation of images is improved by automatically recognising commonly used cardiac magnetic resonance imaging views from image content. The method is based on decision forests and convolutional neural networks trained on a large image dataset. Second, the thesis explores ways to use machine learning for extraction of relevant clinical measures (e.g. volumes and masses) from3D and 3D+t cardiac images. New spatio-temporal image features are designed andclassification forests are trained to learn how to automatically segment the main cardiac structures (left ventricle and left atrium) from voxel-wise label maps. Third, a web interface is designed to collect pairwise image comparisons and to learn how to describe the hearts with semantic attributes (e.g. dilation, kineticity). In the last part of the thesis, a forest-based machinelearning technique is used to map cardiac images to establish distances and neighborhoods between images. One application is retrieval of the most similar images
Vu, Viet-Vu. "Clustering semi-supervisé et apprentissage actif." Paris 6, 2011. http://www.theses.fr/2011PA066607.
Full textDuneau, Laurent. "Etude et réalisation d'un système adaptatif pour la reconnaissance en ligne de mots manuscrits." Compiègne, 1994. http://www.theses.fr/1994COMP7665.
Full textRobinel, Audrey. "Analyse comportementale automatique par des systèmes apprenants." Antilles-Guyane, 2013. http://www.theses.fr/2013AGUY0672.
Full textAlcheikh, Hamoud Khaled. "Modélisation des grands systèmes électriques interconnectés : application à l'analyse de sécurité dans un environnement compétitif." Grenoble INPG, 2010. http://www.theses.fr/2010INPG0032.
Full textZehraoui, Farida. "Systèmes d'apprentissage connexionnistes et raisonnement à partir de cas pour la classification et le classement de séquence." Paris 13, 2004. http://www.theses.fr/2004PA132007.
Full textPaulin, Mathias. "Contributions à l'apprentissage automatique de réseau de contraintes et à la constitution automatique de comportements sensorimoteurs en robotique." Phd thesis, Université Montpellier II - Sciences et Techniques du Languedoc, 2008. http://tel.archives-ouvertes.fr/tel-00340438.
Full textBoucheron, Stéphane. "Apprentissage et calculs." Montpellier 2, 1988. http://www.theses.fr/1988MON20251.
Full textMoneret, Régis. "Strategos : un système multi-jeux utilisant la théorie combinatoire des jeux, capable d'apprendre automatiquement les dépendances entre sous-jeux locaux." Paris 6, 2000. http://www.theses.fr/2000PA066338.
Full textPessiot, Jean-François. "Apprentissage automatique pour l'extraction de caractéristiques : application au partitionnement de documents, au résumé automatique et au filtrage collaboratif." Paris 6, 2008. http://www.theses.fr/2008PA066218.
Full textAkindele, Oluwatoyin Tunde. "Vers un système de construction automatique de modèles génériques de structures de documents." Nancy 1, 1995. http://www.theses.fr/1995NAN10002.
Full textMartin, Arnaud. "Évolution de profils multi-attributs, par apprentissage automatique et adaptatif dans un système de recommandation pour l'aide à la décision." Toulouse 3, 2012. http://thesesups.ups-tlse.fr/1753/.
Full textConsidering user profiles and their evolutions, for decision support is currently in the community of DSS (Decision Support Systems) an important issue. Indeed, the inclusion of context in the decision is currently emerging for DSS. Indeed the system offers advice to users based on their profile, which represents their preferences through a list of valued criteria. The main constraints come from the fact that the system need to continuously bring relevant information. It therefore requires changing user profiles thanks to their actions. So, the system must not only "understand" what the user likes, but also why. The users' assistance will evolve over time and therefore with the user. Thus the user has at his disposal a kind of personal assistant. The objective of this work is to provide assistance to the user's activity according to his profile. The objective is to develop an algorithm based on automatic techniques, in order to change the profile of a user based on his actions. The assistance provided to the user by the system will evolves according to the evolution of its profile. The problem addressed to the user is a problem of decision making. For this problem, assistance is provided to the user, and it is a refinement of potential solutions. This refining is done through the establishment of scalable scheduling solutions that are presented to the user depending on his / her profile. The realization of such a system requires the articulation of the three main areas of research which are the Multi-Criteria Decision Support, the Disaggregation and Aggregation of preferences, and Machine Learning. The fields of Decision Support and Multi Disaggregation and Aggregation preference can also be assembled as Multi-Criteria Aggregation Process (PAMC). Some methods of Multicriteria Decision Support are set up here and use profile data to provide the best possible support to the user. The decomposition is used to characterize an object to provide data to the learning algorithm required for its operation. Aggregation serves to score an object according to the user profile in order to rank the selected items. Machine Learning is used to change user profiles in order to always have a profile representing as closely as possible the preferences of users. Indeed user preferences change over the time, it is necessary to address these changes in order to adapt the answers to the user. The contributions of this thesis are firstly, the definition, construction and evolution of a user profile (evolutionary profiling) based on explicit and implicit user's actions. This evolutionary profiling is implemented within a recommender system usable without learning base, synchronously and completely incremental, and that allows users to quickly change their preferences and even to be inconsistent (bounded rationality). This system, which complements an Information System Research, aims to establish a total order on a list of items proposed to the user (ranking) and in accordance with his preferences. These also include the definition of techniques used to make parts of solutions to technological challenges as the disintegration of criteria and the inclusion of a variable number of criteria in the process of interactive decision support, and this without firstly defining coherent family of criteria on which the decision is based. Several application frameworks have been developed to evaluate the system and compare it to other systems, but also to test its performance with real user data in an offline mode, and in an online mode using directly the system
Martinez, Margarit Aleix. "Apprentissage visuel dans un système de vision active : application dans un contexte de robotique et reconnaissance du visage." Paris 8, 1998. http://www.theses.fr/1998PA081521.
Full textDussart, Claude. "Méta-apprentissage dans un environnement d'expériences distribuées." Lyon 1, 2002. http://www.theses.fr/2002LYO10118.
Full textCrémilleux, Bruno. "Induction automatique : aspects théoriques, le système ARBRE, applications en médecine." Phd thesis, Grenoble 1, 1991. http://tel.archives-ouvertes.fr/tel-00339492.
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 textDaoudlarian, Douglas. "Rôle des interactions entre les systèmes immunitaire et nerveux : études préclinique et clinique." Thesis, Université Côte d'Azur (ComUE), 2018. http://www.theses.fr/2018AZUR4032.
Full textWhile the immune system is well known for its protective role against infectious pathogens, its role in cancer progression is more complex with some immune mechanisms being protective while others are detrimental. The primary physiological role of the brain is to perceive external physical and social conditions, assess their implications for organismal well-being and modulate the activity of internal physiological processes to optimally adapt to those external conditions. Immune and the nervous systems have long been considered to operate independently from each other, many preclinical and clinical studies have clearly demonstrated that these two systems interact and regulate each other. Despite more and more studies aim at investigating the interactions between the nervous and the immune systems, important issues remain to be elucidated. For example, while human studies have demonstrated a positive impact of well-being on cancer progression, the underlying molecular mechanisms have not been elucidated. On another topic, and while many investigators have investigated whether cytokines could be used as diagnosis or prognosis biomarkers is psychiatric diseases, none of the cytokine studied to date have proven to possess the sensitivity and specificity expected for an accepted diagnostic test value. During my PhD, I have worked on two different projects both related to the interactions between the nervous and the immune system. The goal of my first project was to elucidate the mechanisms by which enriched environment conductive to enhanced sensory, cognitive and motor stimulation impact metastatic progression in mice. We have found that mice housed in enriched environment were protected from lung metastasis. Protection was associated with lower serum corticosterone levels, increased lung inflammation following extravasation of circulating tumour cells, and rapid killing of early infiltrating tumour cells. Protection was abolished when inflammatory monocytes were deficient in glucocorticoid receptor signalling. Thus, while inflammatory monocytes have been shown to promote cancer progression, our results disclosed a novel anti-tumour mechanism whereby glucocorticoid receptor-dependent reprogramming of inflammatory monocytes can inhibit cancer metastasis. The goal of my second project was to identify immune-related biomarkers of remission in first-episode psychotic (FEP) patients. To this aim, we have taken advantage of our privileged access to clinical data and serum samples from 325 FEP patients who have all been treated with an atypical antipsychotic. We have first used a hierarchical unsupervised clustering approach to stratify 325 FEP patients into four subtypes based on their clinical symptoms. Compared to the rest of the cohort, one subtype (C1A) exhibited more severe positive and negative symptoms and were the most at risk of being non-remitters following treatment for 4 weeks. C1A patients also exhibited higher levels of several pro-inflammatory biomarkers therefore providing an external validation to our clustering approach. Most importantly, six biological variables (serum levels of IL-15, C reactive protein, CXCL-12, anti- cytomegalovirus and anti-Toxoplasma immunoglobulins) and two clinical variables (age, recreational drug use), predicted early remission following treatment with Amisulpride in C1A patients. Prediction accuracy assessed by cross-validation calculated by 10,000 iterations of 4-fold cross-validation was very good with a mean area under the curve (AUC) of 81.0% ± 0.05. Further validation of our results in future clinical trials would pave the way for the development of a blood-based assisted clinical decision support system for the choice of treatment in psychotic patients
Veillon, Lise-Marie. "Apprentissage artificiel collectif ; aspects dynamiques et structurels." Thesis, Sorbonne Paris Cité, 2018. http://www.theses.fr/2018USPCD004/document.
Full textCollective learning in multi-agent systems considers how a community of autonomous agents sharing a learning purpose may benefit from exchanging information to learn efficiently as a community as well as individuals. The community forms a communication network where each agent may accesses observations, called learning examples. This thesis is based on a former protocol, SMILE (Sound-Multi-agent-Incremental-LEarning), which sets up parsimonious examples and hypotheses exchanges between agents. In a fully connected community, this protocol guarantees an agent’s hypothesis takes into account all the examples obtained by the community. Some sequential protocols add propagation to SMILE in order to extend this consistency guarantee to other connected networks. This thesis contribution to the artificial collective learning field is two fold.First, we investigate the influence of network structures on learning in networks when communication is limited to neighbourhood without further information propagation. Second, we present and analyze a new protocol, Waves, with SMILE’s guarantees and a more dynamic learning process thanks to its execution in parallel. The evaluation of this protocol in a simple turn-based setting gives the opportunity to improve it here in multiple ways. It is however meant to be used with online learning without any restriction on the acquisition rate of new examples, neither on speed nor number
Caillaud, Bertrand. "Apprentissage de connaissances prosodiques pour la reconnaissance automatique de la parole." Grenoble INPG, 1996. http://www.theses.fr/1996INPG0219.
Full textPaulin, Mathias. "Contributions à l'apprentissage automatique de réseau de contraintes et à la constitution automatique de comportements sensorimoteurs en robotique." Phd thesis, Montpellier 2, 2008. http://www.theses.fr/2008MON20064.
Full textIn the first part of this Ph. D. Thesis, we propose an interactive version of the constraint network acquisition platform CONACQ in which the system actively asks questions to the user in order to increase more rapidly and consistently the knowledge of the platform. We propose a number of algorithms for identifying good queries for acquiring constraint networks and our empirical studies show that using our techniques the number of examples required to acquire a constraint network is significantly reduced. In the second part, we are interested in a practical use of the automatic constraint network acquisition in Robotics. Our approach uses CONACQ in order to model automatically the elementary actions of a robot with constraint networks. These are then combined by planning in order to automatically define a sequence of elementary actions which must be executed by the robot to perform a sensorimotor behaviour
Ferret, Olivier. "ANTHAPSI : un système d'analyse thématique et d'apprentissage de connaissances pragmatiques fondé sur l'amorçage." Phd thesis, Université Paris Sud - Paris XI, 1998. http://tel.archives-ouvertes.fr/tel-00189116.
Full textOsório, Fernando Santos. "Inss : un système hybride neuro-symbolique pour l'apprentissage automatique constructif." Grenoble INPG, 1998. https://tel.archives-ouvertes.fr/tel-00004899.
Full textVarious Artificial Intelligence methods have been developed to reproduce intelligent human behaviour. These methods allow to reproduce some human reasoning process using the available knowledge. Each method has its advantages, but also some drawbacks. Hybrid systems combine different approaches in order to take advantage of their respective strengths. These hybrid intelligent systems also present the ability to acquire new knowledge from different sources and so to improve their application performance. This thesis presents our research in the field of hybrid neuro-symbolic systems, and in particular the study of machine learning tools used for constructive knowledge acquisition. We are interested in the automatic acquisition of theoretical knowledge (rules) and empirical knowledge (examples). We present a new hybrid system we implemented: INSS - Incremental Neuro-Symbolic System. This system allows knowledge transfer from the symbolic module to the connectionist module (Artificial Neural Network - ANN), through symbolic rule compilation into an ANN. We can refine the initial ANN knowledge through neural learning using a set of examples. The incremental ANN learning method used, the Cascade-Correlation algorithm, allows us to change or to add new knowledge to the network. Then, the system can also extract modified (or new) symbolic rules from the ANN and validate them. INSS is a hybrid machine learning system that implements a constructive knowledge acquisition method. We conclude by showing the results we obtained with this system in different application domains: ANN artificial problems(The Monk's Problems), computer aided medical diagnosis (Toxic Comas), a cognitive modelling task (The Balance Scale Problem) and autonomous robot control. The results we obtained show the improved performance of INSS and its advantages over others hybrid neuro-symbolic systems
Hadj-Mabrouk, Habib. "Apprentissage automatique et acquisition des connaissances : deux approches complementaires pour les systèmes à base de connaissances : application au système Acasya d'aide à la certification des systèmes de transport automatisés." Valenciennes, 1992. https://ged.uphf.fr/nuxeo/site/esupversions/baee7687-792f-4762-b86e-586d4cbd0596.
Full textRonceray, Lilian. "Méthodologies de réglage automatique temps-réel de lois de pilotage." Phd thesis, Toulouse, ISAE, 2009. http://tel.archives-ouvertes.fr/tel-00430820.
Full textPomorski, Denis. "Apprentissage automatique symbolique/numérique : construction et évaluation d'un ensemble de règles à partir des données." Lille 1, 1991. http://www.theses.fr/1991LIL10117.
Full textZribi, Abir. "Apprentissage par noyaux multiples : application à la classification automatique des images biomédicales microscopiques." Thesis, Rouen, INSA, 2016. http://www.theses.fr/2016ISAM0001.
Full textThis thesis arises in the context of computer aided analysis for subcellular protein localization in microscopic images. The aim is the establishment of an automatic classification system allowing to identify the cellular compartment in which a protein of interest exerts its biological activity. In order to overcome the difficulties in attempting to discern the cellular compartments in microscopic images, the existing state-of-art systems use several descriptors to train an ensemble of classifiers. In this thesis, we propose a different classification scheme wich better cope with the requirement of genericity and flexibility to treat various image datasets. Aiming to provide an efficient image characterization of microscopic images, a new feature system combining local, frequency-domain, global, and region-based features is proposed. Then, we formulate the problem of heterogeneous feature fusion as a kernel selection problem. Using multiple kernel learning, the problems of optimal feature sets selection and classifier training are simultaneously resolved. The proposed combination scheme leads to a simple and a generic framework capable of providing a high performance for microscopy image classification. Extensive experiments were carried out using widely-used and best known datasets. When compared with the state-of-the-art systems, our framework is more generic and outperforms other classification systems. To further expand our study on multiple kernel learning, we introduce a new formalism for learning with multiple kernels performed in two steps. This contribution consists in proposing three regularized terms with in the minimization of kernels weights problem, formulated as a classification problem using Separators with Vast Margin on the space of pairs of data. The first term ensures that kernels selection leads to a sparse representation. While the second and the third terms introduce the concept of kernels similarity by using a correlation measure. Experiments on various biomedical image datasets show a promising performance of our method compared to states of art methods
Grivolla, Jens. "Apprentissage et décision automatique en recherche documentaire : prédiction de difficulté de requêtes et sélection de modèle de recherche." Avignon, 2006. http://www.theses.fr/2006AVIG0142.
Full textThis thesis is centered around the subject of information retrieval, with a focus on those queries that are particularly difficult to handle for current retrieval systems. In the application and evaluation settings we were concerned with, a user expresses his information need as a natural language query. There are different approaches for treating those queries, but current systems typically use a single approach for all queries, without taking into account the specific properties of each query. However, it has been shown that the performance of one strategy relative to another can vary greatly depending on the query. We have approached this problem by proposing methods that will permit to automatically identify those queries that will pose particular difficulties to the retrieval system, in order to allow for a specific treatment. This research topic was very new and barely starting to be explored at the beginning of my work, but has received much attention these last years. We have developed a certain number of quality predictor functions that obtain results comparable to those published recently by other research teams. However, the ability of individual predictors to accurately classify queries by their level of difficulty remains rather limited. The major particularity and originality of our work lies in the combination of those different measures. Using methods of automatic classification with corpus-based training, we have been able to obtain quite reliable predictions, on the basis of measures that individually are far less discriminant. We have also adapted our approach to other application settings, with very encouraging results. We have thus developed a method for the selective application of query expansion techniques, as well as the selection of the most appropriate retrieval model for each query
Sandoz, Françoise. "Contribution à la modélisation de l'apprentissage en situation de résolution de problèmes : proposition d'un système cognitif permettant l'apprentissage." Besançon, 1993. http://www.theses.fr/1993BESA2012.
Full textThomas, Vincent. "Proposition d'un formalisme pour la construction automatique d'interactions dans les systèmes multi-agents réactifs." Phd thesis, Université Henri Poincaré - Nancy I, 2005. http://tel.archives-ouvertes.fr/tel-00011094.
Full textLes formalismes existants comme les DEC-POMDPs parviennent à représenter des problèmes multi-agents mais ne représentent pas au niveau individuel la notion d'interaction fondamentale dans les systèmes collectifs. Ceci induit une complexité algorithmique importante dans les algorithmes de résolution. Afin de donner aux agents la possibilité d'appréhender la présence d'autres agents et de structurer de manière implicite les systèmes multi-agents, cette thèse propose un formalisme original, l'interac-DEC-POMDP inspiré des DEC-POMDPs et d'Hamelin, une simulation développée au cours de cette thèse et issue d'expériences conduites en éthologie. La spécificité de ce formalisme réside dans la capacité offerte aux agents d'interagir directement et localement entre eux. Cette possibilité permet des prises de décision à un niveau intermédiaire entre des décisions globales impliquant l'ensemble des agents et des décisions purement individuelles.
Nous avons proposé en outre un algorithme décentralisé basé sur des techniques d'apprentissage par renforcement et une répartition heuristique des gains des agents au cours des interactions. Une démarche expérimentale nous a permis de valider sa capacité à produire pour des restriction du formalisme des comportements collectifs pertinents adaptatifs sans qu'aucun agent ne dispose d'une vue globale du système.
Makiou, Abdelhamid. "Sécurité des applications Web : Analyse, modélisation et détection des attaques par apprentissage automatique." Thesis, Paris, ENST, 2016. http://www.theses.fr/2016ENST0084/document.
Full textWeb applications are the backbone of modern information systems. The Internet exposure of these applications continually generates new forms of threats that can jeopardize the security of the entire information system. To counter these threats, there are robust and feature-rich solutions. These solutions are based on well-proven attack detection models, with advantages and limitations for each model. Our work consists in integrating functionalities of several models into a single solution in order to increase the detection capacity. To achieve this objective, we define in a first contribution, a classification of the threats adapted to the context of the Web applications. This classification also serves to solve some problems of scheduling analysis operations during the detection phase of the attacks. In a second contribution, we propose an architecture of Web application firewall based on two analysis models. The first is a behavioral analysis module, and the second uses the signature inspection approach. The main challenge to be addressed with this architecture is to adapt the behavioral analysis model to the context of Web applications. We are responding to this challenge by using a modeling approach of malicious behavior. Thus, it is possible to construct for each attack class its own model of abnormal behavior. To construct these models, we use classifiers based on supervised machine learning. These classifiers use learning datasets to learn the deviant behaviors of each class of attacks. Thus, a second lock in terms of the availability of the learning data has been lifted. Indeed, in a final contribution, we defined and designed a platform for automatic generation of training datasets. The data generated by this platform is standardized and categorized for each class of attacks. The learning data generation model we have developed is able to learn "from its own errors" continuously in order to produce higher quality machine learning datasets
El, Hatib Souad. "Une approche sémantique de détection de maliciel Android basée sur la vérification de modèles et l'apprentissage automatique." Master's thesis, Université Laval, 2020. http://hdl.handle.net/20.500.11794/66322.
Full textThe ever-increasing number of Android malware is accompanied by a deep concern about security issues in the mobile ecosystem. Unquestionably, Android malware detection has received much attention in the research community and therefore it becomes a crucial aspect of software security. Actually, malware proliferation goes hand in hand with the sophistication and complexity of malware. To illustrate, more elaborated malware like polymorphic and metamorphic malware, make use of code obfuscation techniques to build new variants that preserve the semantics of the original code but modify it’s syntax and thus escape the usual detection methods. In the present work, we propose a model-checking based approach that combines static analysis and machine learning. Mainly, from a given Android application we extract an abstract model expressed in terms of LNT, a process algebra language. Afterwards, security related Android behaviours specified by temporal logic formulas are checked against this model, the satisfaction of a specific formula is considered as a feature, finally machine learning algorithms are used to classify the application as malicious or not.
Halluin, Cyrille d'. "Apprentissage PAC par exemples simples : plate-forme d'apprentissage de langages réguliers." Lille 1, 1998. https://pepite-depot.univ-lille.fr/LIBRE/Th_Num/1998/50376-1998-105.pdf.
Full textServan, Christophe. "Apprentissage automatique et compréhension dans le cadre d'un dialogue homme-machine téléphonique à initiative mixte." Phd thesis, Université d'Avignon, 2008. http://tel.archives-ouvertes.fr/tel-00591997.
Full textPradel, 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
Dupas, Rémy. "Apport des méthodes d'apprentissage symbolique automatique pour l'aide à la maintenance industrielle." Valenciennes, 1990. https://ged.uphf.fr/nuxeo/site/esupversions/7ab53b01-cdfb-4932-ba60-cb5332e3925a.
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
Kraus, Vivien. "Apprentissage semi-supervisé pour la régression multi-labels : application à l’annotation automatique de pneumatiques." Thesis, Lyon, 2021. https://tel.archives-ouvertes.fr/tel-03789608.
Full textWith the advent and rapid growth of digital technologies, data has become a precious asset as well as plentiful. However, with such an abundance come issues about data quality and labelling. Because of growing numbers of available data volumes, while human expert labelling is still important, it is more and more necessary to reinforce semi-supervised learning with the exploitation of unlabeled data. This problem is all the more noticeable in the multi-label learning framework, and in particular for regression, where each statistical unit is guided by many different targets, taking the form of numerical scores. This thesis focuses on this fundamental framework. First, we begin by proposing a method for semi-supervised regression, that we challenge through a detailed experimental study. Thanks to this new method, we present a second contribution, more fitted to the multi-label framework. We also show its efficiency with a comparative study on literature data sets. Furthermore, the problem dimension is always a pain point of machine learning, and reducing it sparks the interest of many researchers. Feature selection is one of the major tasks addressing this problem, and we propose to study it here in a complex framework : for semi-supervised, multi-label regression. Finally, an experimental validation is proposed on a real problem about automatic annotation of tires, to tackle the needs expressed by the industrial partner of this thesis
Olaru, Andrei. "Un système multi-agents sensible au contexte pour les environnements d'intelligence ambiante." Paris 6, 2011. http://www.theses.fr/2011PA066639.
Full text