Дисертації з теми "Apprentissage profond par renforcement"
Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями
Ознайомтеся з топ-50 дисертацій для дослідження на тему "Apprentissage profond par renforcement".
Біля кожної праці в переліку літератури доступна кнопка «Додати до бібліографії». Скористайтеся нею – і ми автоматично оформимо бібліографічне посилання на обрану працю в потрібному вам стилі цитування: APA, MLA, «Гарвард», «Чикаго», «Ванкувер» тощо.
Також ви можете завантажити повний текст наукової публікації у форматі «.pdf» та прочитати онлайн анотацію до роботи, якщо відповідні параметри наявні в метаданих.
Переглядайте дисертації для різних дисциплін та оформлюйте правильно вашу бібліографію.
Zimmer, Matthieu. "Apprentissage par renforcement développemental." Thesis, Université de Lorraine, 2018. http://www.theses.fr/2018LORR0008/document.
Повний текст джерелаReinforcement learning allows an agent to learn a behavior that has never been previously defined by humans. The agent discovers the environment and the different consequences of its actions through its interaction: it learns from its own experience, without having pre-established knowledge of the goals or effects of its actions. This thesis tackles how deep learning can help reinforcement learning to handle continuous spaces and environments with many degrees of freedom in order to solve problems closer to reality. Indeed, neural networks have a good scalability and representativeness. They make possible to approximate functions on continuous spaces and allow a developmental approach, because they require little a priori knowledge on the domain. We seek to reduce the amount of necessary interaction of the agent to achieve acceptable behavior. To do so, we proposed the Neural Fitted Actor-Critic framework that defines several data efficient actor-critic algorithms. We examine how the agent can fully exploit the transitions generated by previous behaviors by integrating off-policy data into the proposed framework. Finally, we study how the agent can learn faster by taking advantage of the development of his body, in particular, by proceeding with a gradual increase in the dimensionality of its sensorimotor space
Martinez, Coralie. "Classification précoce de séquences temporelles par de l'apprentissage par renforcement profond." Thesis, Université Grenoble Alpes (ComUE), 2019. http://www.theses.fr/2019GREAT123.
Повний текст джерелаEarly classification (EC) of time series is a recent research topic in the field of sequential data analysis. It consists in assigning a label to some data that is sequentially collected with new data points arriving over time, and the prediction of a label has to be made using as few data points as possible in the sequence. The EC problem is of paramount importance for supporting decision makers in many real-world applications, ranging from process control to fraud detection. It is particularly interesting for applications concerned with the costs induced by the acquisition of data points, or for applications which seek for rapid label prediction in order to take early actions. This is for example the case in the field of health, where it is necessary to provide a medical diagnosis as soon as possible from the sequence of medical observations collected over time. Another example is predictive maintenance with the objective to anticipate the breakdown of a machine from its sensor signals. In this doctoral work, we developed a new approach for this problem, based on the formulation of a sequential decision making problem, that is the EC model has to decide between classifying an incomplete sequence or delaying the prediction to collect additional data points. Specifically, we described this problem as a Partially Observable Markov Decision Process noted EC-POMDP. The approach consists in training an EC agent with Deep Reinforcement Learning (DRL) in an environment characterized by the EC-POMDP. The main motivation for this approach was to offer an end-to-end model for EC which is able to simultaneously learn optimal patterns in the sequences for classification and optimal strategic decisions for the time of prediction. Also, the method allows to set the importance of time against accuracy of the classification in the definition of rewards, according to the application and its willingness to make this compromise. In order to solve the EC-POMDP and model the policy of the EC agent, we applied an existing DRL algorithm, the Double Deep-Q-Network algorithm, whose general principle is to update the policy of the agent during training episodes, using a replay memory of past experiences. We showed that the application of the original algorithm to the EC problem lead to imbalanced memory issues which can weaken the training of the agent. Consequently, to cope with those issues and offer a more robust training of the agent, we adapted the algorithm to the EC-POMDP specificities and we introduced strategies of memory management and episode management. In experiments, we showed that these contributions improved the performance of the agent over the original algorithm, and that we were able to train an EC agent which compromised between speed and accuracy, on each sequence individually. We were also able to train EC agents on public datasets for which we have no expertise, showing that the method is applicable to various domains. Finally, we proposed some strategies to interpret the decisions of the agent, validate or reject them. In experiments, we showed how these solutions can help gain insight in the choice of action made by the agent
Paumard, Marie-Morgane. "Résolution automatique de puzzles par apprentissage profond." Thesis, CY Cergy Paris Université, 2020. http://www.theses.fr/2020CYUN1067.
Повний текст джерелаThe objective of this thesis is to develop semantic methods of reassembly in the complicated framework of heritage collections, where some blocks are eroded or missing.The reassembly of archaeological remains is an important task for heritage sciences: it allows to improve the understanding and conservation of ancient vestiges and artifacts. However, some sets of fragments cannot be reassembled with techniques using contour information or visual continuities. It is then necessary to extract semantic information from the fragments and to interpret them. These tasks can be performed automatically thanks to deep learning techniques coupled with a solver, i.e., a constrained decision making algorithm.This thesis proposes two semantic reassembly methods for 2D fragments with erosion and a new dataset and evaluation metrics.The first method, Deepzzle, proposes a neural network followed by a solver. The neural network is composed of two Siamese convolutional networks trained to predict the relative position of two fragments: it is a 9-class classification. The solver uses Dijkstra's algorithm to maximize the joint probability. Deepzzle can address the case of missing and supernumerary fragments, is capable of processing about 15 fragments per puzzle, and has a performance that is 25% better than the state of the art.The second method, Alphazzle, is based on AlphaZero and single-player Monte Carlo Tree Search (MCTS). It is an iterative method that uses deep reinforcement learning: at each step, a fragment is placed on the current reassembly. Two neural networks guide MCTS: an action predictor, which uses the fragment and the current reassembly to propose a strategy, and an evaluator, which is trained to predict the quality of the future result from the current reassembly. Alphazzle takes into account the relationships between all fragments and adapts to puzzles larger than those solved by Deepzzle. Moreover, Alphazzle is compatible with constraints imposed by a heritage framework: at the end of reassembly, MCTS does not access the reward, unlike AlphaZero. Indeed, the reward, which indicates if a puzzle is well solved or not, can only be estimated by the algorithm, because only a conservator can be sure of the quality of a reassembly
Léon, Aurélia. "Apprentissage séquentiel budgétisé pour la classification extrême et la découverte de hiérarchie en apprentissage par renforcement." Electronic Thesis or Diss., Sorbonne université, 2019. http://www.theses.fr/2019SORUS226.
Повний текст джерелаThis thesis deals with the notion of budget to study problems of complexity (it can be computational complexity, a complex task for an agent, or complexity due to a small amount of data). Indeed, the main goal of current techniques in machine learning is usually to obtain the best accuracy, without worrying about the cost of the task. The concept of budget makes it possible to take into account this parameter while maintaining good performances. We first focus on classification problems with a large number of classes: the complexity in those algorithms can be reduced thanks to the use of decision trees (here learned through budgeted reinforcement learning techniques) or the association of each class with a (binary) code. We then deal with reinforcement learning problems and the discovery of a hierarchy that breaks down a (complex) task into simpler tasks to facilitate learning and generalization. Here, this discovery is done by reducing the cognitive effort of the agent (considered in this work as equivalent to the use of an additional observation). Finally, we address problems of understanding and generating instructions in natural language, where data are available in small quantities: we test for this purpose the simultaneous use of an agent that understands and of an agent that generates the instructions
Brenon, Alexis. "Modèle profond pour le contrôle vocal adaptatif d'un habitat intelligent." Thesis, Université Grenoble Alpes (ComUE), 2017. http://www.theses.fr/2017GREAM057/document.
Повний текст джерелаSmart-homes, resulting of the merger of home-automation, ubiquitous computing and artificial intelligence, support inhabitants in their activity of daily living to improve their quality of life.Allowing dependent and aged people to live at home longer, these homes provide a first answer to society problems as the dependency tied to the aging population.In voice controlled home, the home has to answer to user's requests covering a range of automated actions (lights, blinds, multimedia control, etc.).To achieve this, the control system of the home need to be aware of the context in which a request has been done, but also to know user habits and preferences.Thus, the system must be able to aggregate information from a heterogeneous home-automation sensors network and take the (variable) user behavior into account.The development of smart home control systems is hard due to the huge variability regarding the home topology and the user habits.Furthermore, the whole set of contextual information need to be represented in a common space in order to be able to reason about them and make decisions.To address these problems, we propose to develop a system which updates continuously its model to adapt itself to the user and which uses raw data from the sensors through a graphical representation.This new method is particularly interesting because it does not require any prior inference step to extract the context.Thus, our system uses deep reinforcement learning; a convolutional neural network allowing to extract contextual information and reinforcement learning used for decision-making.Then, this memoir presents two systems, a first one only based on reinforcement learning showing limits of this approach against real environment with thousands of possible states.Introduction of deep learning allowed to develop the second one, ARCADES, which gives good performances proving that this approach is relevant and opening many ways to improve it
Carrara, Nicolas. "Reinforcement learning for dialogue systems optimization with user adaptation." Thesis, Lille 1, 2019. http://www.theses.fr/2019LIL1I071/document.
Повний текст джерелаThe most powerful artificial intelligence systems are now based on learned statistical models. In order to build efficient models, these systems must collect a huge amount of data on their environment. Personal assistants, smart-homes, voice-servers and other dialogue applications are no exceptions to this statement. A specificity of those systems is that they are designed to interact with humans, and as a consequence, their training data has to be collected from interactions with these humans. As the number of interactions with a single person is often too scarce to train a proper model, the usual approach to maximise the amount of data consists in mixing data collected with different users into a single corpus. However, one limitation of this approach is that, by construction, the trained models are only efficient with an "average" human and do not include any sort of adaptation; this lack of adaptation makes the service unusable for some specific group of persons and leads to a restricted customers base and inclusiveness problems. This thesis proposes solutions to construct Dialogue Systems that are robust to this problem by combining Transfer Learning and Reinforcement Learning. It explores two main ideas: The first idea of this thesis consists in incorporating adaptation in the very first dialogues with a new user. To that extend, we use the knowledge gathered with previous users. But how to scale such systems with a growing database of user interactions? The first proposed approach involves clustering of Dialogue Systems (tailored for their respective user) based on their behaviours. We demonstrated through handcrafted and real user-models experiments how this method improves the dialogue quality for new and unknown users. The second approach extends the Deep Q-learning algorithm with a continuous transfer process.The second idea states that before using a dedicated Dialogue System, the first interactions with a user should be handled carefully by a safe Dialogue System common to all users. The underlying approach is divided in two steps. The first step consists in learning a safe strategy through Reinforcement Learning. To that extent, we introduced a budgeted Reinforcement Learning framework for continuous state space and the underlying extensions of classic Reinforcement Learning algorithms. In particular, the safe version of the Fitted-Q algorithm has been validated, in term of safety and efficiency, on a dialogue system tasks and an autonomous driving problem. The second step consists in using those safe strategies when facing new users; this method is an extension of the classic ε-greedy algorithm
Aklil, Nassim. "Apprentissage actif sous contrainte de budget en robotique et en neurosciences computationnelles. Localisation robotique et modélisation comportementale en environnement non stationnaire." Thesis, Paris 6, 2017. http://www.theses.fr/2017PA066225/document.
Повний текст джерелаDecision-making is a highly researched field in science, be it in neuroscience to understand the processes underlying animal decision-making, or in robotics to model efficient and rapid decision-making processes in real environments. In neuroscience, this problem is resolved online with sequential decision-making models based on reinforcement learning. In robotics, the primary objective is efficiency, in order to be deployed in real environments. However, in robotics what can be called the budget and which concerns the limitations inherent to the hardware, such as computation times, limited actions available to the robot or the lifetime of the robot battery, are often not taken into account at the present time. We propose in this thesis to introduce the notion of budget as an explicit constraint in the robotic learning processes applied to a localization task by implementing a model based on work developed in statistical learning that processes data under explicit constraints, limiting the input of data or imposing a more explicit time constraint. In order to discuss an online functioning of this type of budgeted learning algorithms, we also discuss some possible inspirations that could be taken on the side of computational neuroscience. In this context, the alternation between information retrieval for location and the decision to move for a robot may be indirectly linked to the notion of exploration-exploitation compromise. We present our contribution to the modeling of this compromise in animals in a non-stationary task involving different levels of uncertainty, and we make the link with the methods of multi-armed bandits
De, La Bourdonnaye François. "Learning sensori-motor mappings using little knowledge : application to manipulation robotics." Thesis, Université Clermont Auvergne (2017-2020), 2018. http://www.theses.fr/2018CLFAC037/document.
Повний текст джерелаThe thesis is focused on learning a complex manipulation robotics task using little knowledge. More precisely, the concerned task consists in reaching an object with a serial arm and the objective is to learn it without camera calibration parameters, forward kinematics, handcrafted features, or expert demonstrations. Deep reinforcement learning algorithms suit well to this objective. Indeed, reinforcement learning allows to learn sensori-motor mappings while dispensing with dynamics. Besides, deep learning allows to dispense with handcrafted features for the state spacerepresentation. However, it is difficult to specify the objectives of the learned task without requiring human supervision. Some solutions imply expert demonstrations or shaping rewards to guiderobots towards its objective. The latter is generally computed using forward kinematics and handcrafted visual modules. Another class of solutions consists in decomposing the complex task. Learning from easy missions can be used, but this requires the knowledge of a goal state. Decomposing the whole complex into simpler sub tasks can also be utilized (hierarchical learning) but does notnecessarily imply a lack of human supervision. Alternate approaches which use several agents in parallel to increase the probability of success can be used but are costly. In our approach,we decompose the whole reaching task into three simpler sub tasks while taking inspiration from the human behavior. Indeed, humans first look at an object before reaching it. The first learned task is an object fixation task which is aimed at localizing the object in the 3D space. This is learned using deep reinforcement learning and a weakly supervised reward function. The second task consists in learning jointly end-effector binocular fixations and a hand-eye coordination function. This is also learned using a similar set-up and is aimed at localizing the end-effector in the 3D space. The third task uses the two prior learned skills to learn to reach an object and uses the same requirements as the two prior tasks: it hardly requires supervision. In addition, without using additional priors, an object reachability predictor is learned in parallel. The main contribution of this thesis is the learning of a complex robotic task with weak supervision
Pageaud, Simon. "SmartGov : architecture générique pour la co-construction de politiques urbaines basée sur l'apprentissage par renforcement multi-agent." Thesis, Lyon, 2019. http://www.theses.fr/2019LYSE1128.
Повний текст джерелаIn this thesis, we propose the SmartGov model, coupling multi-agent simulation and multi-agent deep reinforcement learning, to help co-construct urban policies and integrate all stakeholders in the decision process. Smart Cities provide sensor data from the urban areas to increase realism of the simulation in SmartGov.Our first contribution is a generic architecture for multi-agent simulation of the city to study global behavior emergence with realistic agents reacting to political decisions. With a multi-level modeling and a coupling of different dynamics, our tool learns environment specificities and suggests relevant policies. Our second contribution improves autonomy and adaptation of the decision function with multi-agent, multi-level reinforcement learning. A set of clustered agents is distributed over the studied area to learn local specificities without any prior knowledge on the environment. Trust score assignment and individual rewards help reduce non-stationary impact on experience replay in deep reinforcement learning.These contributions bring forth a complete system to co-construct urban policies in the Smart City. We compare our model with different approaches from the literature on a parking fee policy to display the benefits and limits of our contributions
Debard, Quentin. "Automatic learning of next generation human-computer interactions." Thesis, Lyon, 2020. http://www.theses.fr/2020LYSEI036.
Повний текст джерелаArtificial Intelligence (AI) and Human-Computer Interactions (HCIs) are two research fields with relatively few common work. HCI specialists usually design the way we interact with devices directly from observations and measures of human feedback, manually optimizing the user interface to better fit users’ expectations. This process is hard to optimize: ergonomy, intuitivity and ease of use are key features in a User Interface (UI) that are too complex to be simply modelled from interaction data. This drastically restrains the possible uses of Machine Learning (ML) in this design process. Currently, ML in HCI is mostly applied to gesture recognition and automatic display, e.g. advertisement or item suggestion. It is also used to fine tune an existing UI to better optimize it, but as of now it does not participate in designing new ways to interact with computers. Our main focus in this thesis is to use ML to develop new design strategies for overall better UIs. We want to use ML to build intelligent – understand precise, intuitive and adaptive – user interfaces using minimal handcrafting. We propose a novel approach to UI design: instead of letting the user adapt to the interface, we want the interface and the user to adapt mutually to each other. The goal is to reduce human bias in protocol definition while building co-adaptive interfaces able to further fit individual preferences. In order to do so, we will put to use the different mechanisms available in ML to automatically learn behaviors, build representations and take decisions. We will be experimenting on touch interfaces, as these interfaces are vastly used and can provide easily interpretable problems. The very first part of our work will focus on processing touch data and use supervised learning to build accurate classifiers of touch gestures. The second part will detail how Reinforcement Learning (RL) can be used to model and learn interaction protocols given user actions. Lastly, we will combine these RL models with unsupervised learning to build a setup allowing for the design of new interaction protocols without the need for real user data
Strub, Florian. "Développement de modèles multimodaux interactifs pour l'apprentissage du langage dans des environnements visuels." Thesis, Lille 1, 2020. http://www.theses.fr/2020LIL1I030.
Повний текст джерелаWhile our representation of the world is shaped by our perceptions, our languages, and our interactions, they have traditionally been distinct fields of study in machine learning. Fortunately, this partitioning started opening up with the recent advents of deep learning methods, which standardized raw feature extraction across communities. However, multimodal neural architectures are still at their beginning, and deep reinforcement learning is often limited to constrained environments. Yet, we ideally aim to develop large-scale multimodal and interactive models towards correctly apprehending the complexity of the world. As a first milestone, this thesis focuses on visually grounded language learning for three reasons (i) they are both well-studied modalities across different scientific fields (ii) it builds upon deep learning breakthroughs in natural language processing and computer vision (ii) the interplay between language and vision has been acknowledged in cognitive science. More precisely, we first designed the GuessWhat?! game for assessing visually grounded language understanding of the models: two players collaborate to locate a hidden object in an image by asking a sequence of questions. We then introduce modulation as a novel deep multimodal mechanism, and we show that it successfully fuses visual and linguistic representations by taking advantage of the hierarchical structure of neural networks. Finally, we investigate how reinforcement learning can support visually grounded language learning and cement the underlying multimodal representation. We show that such interactive learning leads to consistent language strategies but gives raise to new research issues
Sun-Hosoya, Lisheng. "Meta-Learning as a Markov Decision Process." Thesis, Université Paris-Saclay (ComUE), 2019. http://www.theses.fr/2019SACLS588/document.
Повний текст джерелаMachine Learning (ML) has enjoyed huge successes in recent years and an ever- growing number of real-world applications rely on it. However, designing promising algorithms for a specific problem still requires huge human effort. Automated Machine Learning (AutoML) aims at taking the human out of the loop and develop machines that generate / recommend good algorithms for a given ML tasks. AutoML is usually treated as an algorithm / hyper-parameter selection problems, existing approaches include Bayesian optimization, evolutionary algorithms as well as reinforcement learning. Among them, auto-sklearn which incorporates meta-learning techniques in their search initialization, ranks consistently well in AutoML challenges. This observation oriented my research to the Meta-Learning domain. This direction led me to develop a novel framework based on Markov Decision Processes (MDP) and reinforcement learning (RL).After a general introduction (Chapter 1), my thesis work starts with an in-depth analysis of the results of the AutoML challenge (Chapter 2). This analysis oriented my work towards meta-learning, leading me first to propose a formulation of AutoML as a recommendation problem, and ultimately to formulate a novel conceptualisation of the problem as a MDP (Chapter 3). In the MDP setting, the problem is brought back to filling up, as quickly and efficiently as possible, a meta-learning matrix S, in which lines correspond to ML tasks and columns to ML algorithms. A matrix element S(i, j) is the performance of algorithm j applied to task i. Searching efficiently for the best values in S allows us to identify quickly algorithms best suited to given tasks. In Chapter 4 the classical hyper-parameter optimization framework (HyperOpt) is first reviewed. In Chapter 5 a first meta-learning approach is introduced along the lines of our paper ActivMetaL that combines active learning and collaborative filtering techniques to predict the missing values in S. Our latest research applies RL to the MDP problem we defined to learn an efficient policy to explore S. We call this approach REVEAL and propose an analogy with a series of toy games to help visualize agents’ strategies to reveal information progressively, e.g. masked areas of images to be classified, or ship positions in a battleship game. This line of research is developed in Chapter 6. The main results of my PhD project are: 1) HP / model selection: I have explored the Freeze-Thaw method and optimized the algorithm to enter the first AutoML challenge, achieving 3rd place in the final round (Chapter 3). 2) ActivMetaL: I have designed a new algorithm for active meta-learning (ActivMetaL) and compared it with other baseline methods on real-world and artificial data. This study demonstrated that ActiveMetaL is generally able to discover the best algorithm faster than baseline methods. 3) REVEAL: I developed a new conceptualization of meta-learning as a Markov Decision Process and put it into the more general framework of REVEAL games. With a master student intern, I developed agents that learns (with reinforcement learning) to predict the next best algorithm to be tried. To develop this agent, we used surrogate toy tasks of REVEAL games. We then applied our methods to AutoML problems. The work presented in my thesis is empirical in nature. Several real world meta-datasets were used in this research. Artificial and semi-artificial meta-datasets are also used in my work. The results indicate that RL is a viable approach to this problem, although much work remains to be done to optimize algorithms to make them scale to larger meta-learning problems
Choffin, Benoît. "Algorithmes d’espacement adaptatif de l’apprentissage pour l’optimisation de la maîtrise à long terme de composantes de connaissance." Electronic Thesis or Diss., université Paris-Saclay, 2021. http://www.theses.fr/2021UPASG001.
Повний текст джерелаBetween acquiring new knowledge and reviewing old knowledge to mitigate forgetting, learners may find it difficult to organize their learning time effectively. Adaptive spacing algorithms, like SuperMemo, can help learners deal with this trade-off. Such algorithms sequentially plan reviews of a given piece of knowledge to adapt to the specific and ongoing needs of each learner. Compared to a fixed and identical temporal spacing between reviews, several experiments have shown that adaptive spacing improves long-term memory retention of the piece of knowledge.To date, research on adaptive spacing algorithms has focused on the pure memorization of simple pieces of knowledge, which are often represented by flashcards. However, several studies in cognitive psychology have shown that the benefits of spacing out learning episodes on long-term retention also extend to more complex knowledge, such as learning concepts and procedures in mathematics. In this thesis, we have therefore sought to develop adaptive and personalized spacing algorithms for optimizing long-term mastery of knowledge components (KCs).First, we develop and present a new statistical model of learning and forgetting of knowledge components, coined DAS3H, and we empirically show that DAS3H has better predictive performance than several learner models in educational data mining. Second, we develop several adaptive spacing heuristics for long-term mastery of KCs and compare their performance on simulated data. Two of these heuristics use the DAS3H model to select which KC should be reviewed by a given learner at a given time. In addition, we propose a new greedy procedure to select the most promising subset of KCs instead of the best KC to review. Finally, in the last chapter of this thesis, we develop AC4S, a deep reinforcement learning algorithm for adaptive spacing for KCs. We compare this data-driven approach to the heuristic methods that we presented previously
Kozlova, Olga. "Apprentissage par renforcement hiérarchique et factorisé." Phd thesis, Université Pierre et Marie Curie - Paris VI, 2010. http://tel.archives-ouvertes.fr/tel-00632968.
Повний текст джерелаFilippi, Sarah. "Stratégies optimistes en apprentissage par renforcement." Phd thesis, Ecole nationale supérieure des telecommunications - ENST, 2010. http://tel.archives-ouvertes.fr/tel-00551401.
Повний текст джерелаThéro, Héloïse. "Contrôle, agentivité et apprentissage par renforcement." Thesis, Paris Sciences et Lettres (ComUE), 2018. http://www.theses.fr/2018PSLEE028/document.
Повний текст джерелаSense of agency or subjective control can be defined by the feeling that we control our actions, and through them effects in the outside world. This cluster of experiences depend on the ability to learn action-outcome contingencies and a more classical algorithm to model this originates in the field of human reinforcementlearning. In this PhD thesis, we used the cognitive modeling approach to investigate further the interaction between perceived control and reinforcement learning. First, we saw that participants undergoing a reinforcement-learning task experienced higher agency; this influence of reinforcement learning on agency comes as no surprise, because reinforcement learning relies on linking a voluntary action and its outcome. But our results also suggest that agency influences reinforcement learning in two ways. We found that people learn actionoutcome contingencies based on a default assumption: their actions make a difference to the world. Finally, we also found that the mere fact of choosing freely shapes the learning processes following that decision. Our general conclusion is that agency and reinforcement learning, two fundamental fields of human psychology, are deeply intertwined. Contrary to machines, humans do care about being in control, or about making the right choice, and this results in integrating information in a one-sided way
Sola, Yoann. "Contributions to the development of deep reinforcement learning-based controllers for AUV." Thesis, Brest, École nationale supérieure de techniques avancées Bretagne, 2021. http://www.theses.fr/2021ENTA0015.
Повний текст джерелаThe marine environment is a very hostile setting for robotics. It is strongly unstructured, very uncertain and includes a lot of external disturbances which cannot be easily predicted or modelled. In this work, we will try to control an autonomous underwater vehicle (AUV) in order to perform a waypoint tracking task, using a machine learning-based controller. Machine learning allowed to make impressive progress in a lot of different domain in the recent years, and the subfield of deep reinforcement learning managed to design several algorithms very suitable for the continuous control of dynamical systems. We chose to implement the Soft Actor-Critic (SAC) algorithm, an entropy-regularized deep reinforcement learning algorithm allowing to fulfill a learning task and to encourage the exploration of the environment simultaneously. We compared a SAC-based controller with a Proportional-Integral-Derivative (PID) controller on a waypoint tracking task and using specific performance metrics. All the tests were performed in simulation thanks to the use of the UUV Simulator. We decided to apply these two controllers to the RexROV 2, a six degrees of freedom cube-shaped remotely operated underwater vehicle (ROV) converted in an AUV. Thanks to these tests, we managed to propose several interesting contributions such as making the SAC achieve an end-to-end control of the AUV, outperforming the PID controller in terms of energy saving, and reducing the amount of information needed by the SAC algorithm. Moreover we propose a methodology for the training of deep reinforcement learning algorithms on control tasks, as well as a discussion about the absence of guidance algorithms for our end-to-end AUV controller
Munos, Rémi. "Apprentissage par renforcement, étude du cas continu." Paris, EHESS, 1997. http://www.theses.fr/1997EHESA021.
Повний текст джерелаDeschaintre, Valentin. "Acquisition légère de matériaux par apprentissage profond." Thesis, Université Côte d'Azur (ComUE), 2019. http://theses.univ-cotedazur.fr/2019AZUR4078.
Повний текст джерелаWhether it is used for entertainment or industrial design, computer graphics is ever more present in our everyday life. Yet, reproducing a real scene appearance in a virtual environment remains a challenging task, requiring long hours from trained artists. A good solution is the acquisition of geometries and materials directly from real world examples, but this often comes at the cost of complex hardware and calibration processes. In this thesis, we focus on lightweight material appearance capture to simplify and accelerate the acquisition process and solve industrial challenges such as result image resolution or calibration. Texture, highlights, and shading are some of many visual cues that allow humans to perceive material appearance in pictures. Designing algorithms able to leverage these cues to recover spatially-varying bi-directional reflectance distribution functions (SVBRDFs) from a few images has challenged computer graphics researchers for decades. We explore the use of deep learning to tackle lightweight appearance capture and make sense of these visual cues. Once trained, our networks are capable of recovering per-pixel normals, diffuse albedo, specular albedo and specular roughness from as little as one picture of a flat surface lit by the environment or a hand-held flash. We show how our method improves its prediction with the number of input pictures to reach high quality reconstructions with up to 10 images --- a sweet spot between existing single-image and complex multi-image approaches --- and allows to capture large scale, HD materials. We achieve this goal by introducing several innovations on training data acquisition and network design, bringing clear improvement over the state of the art for lightweight material capture
Haykal, Vanessa. "Modélisation des séries temporelles par apprentissage profond." Thesis, Tours, 2019. http://www.theses.fr/2019TOUR4019.
Повний текст джерелаTime series prediction is a problem that has been addressed for many years. In this thesis, we have been interested in methods resulting from deep learning. It is well known that if the relationships between the data are temporal, it is difficult to analyze and predict accurately due to non-linear trends and the existence of noise specifically in the financial and electrical series. From this context, we propose a new hybrid noise reduction architecture that models the recursive error series to improve predictions. The learning process fusessimultaneouslyaconvolutionalneuralnetwork(CNN)andarecurrentlongshort-term memory network (LSTM). This model is distinguished by its ability to capture globally a variety of hybrid properties, where it is able to extract local signal features, to learn long-term and non-linear dependencies, and to have a high noise resistance. The second contribution concerns the limitations of the global approaches because of the dynamic switching regimes in the signal. We present a local unsupervised modification with our previous architecture in order to adjust the results by adapting the Hidden Markov Model (HMM). Finally, we were also interested in multi-resolution techniques to improve the performance of the convolutional layers, notably by using the variational mode decomposition method (VMD)
Ostertag, Cécilia. "Analyse des pathologies neuro-dégénératives par apprentissage profond." Thesis, La Rochelle, 2022. http://www.theses.fr/2022LAROS003.
Повний текст джерелаMonitoring and predicting the cognitive state of a subject affected by a neuro-degenerative disorder is crucial to provide appropriate treatment as soon as possible. Thus, these patients are followed for several years, as part of longitudinal medical studies. During each visit, a large quantity of data is acquired : risk factors linked to the pathology, medical imagery (MRI or PET scans for example), cognitive tests results, sampling of molecules that have been identified as bio-markers, etc. These various modalities give information about the disease's progression, some of them are complementary and others can be redundant. Several deep learning models have been applied to bio-medical data, notably for organ segmentation or pathology diagnosis. This PhD is focused on the conception of a deep neural network model for cognitive decline prediction, using multimodal data, here both structural brain MRI images and clinical data. In this thesis we propose an architecture made of sub-modules tailored to each modality : 3D convolutional network for the brain MRI, and fully connected layers for the quantitative and qualitative clinical data. To predict the patient's evolution, this model takes as input data from two medical visits for each patient. These visits are compared using a siamese architecture. After training and validating this model with Alzheimer's disease as our use case, we look into knowledge transfer to other neuro-degenerative pathologies, and we use transfer learning to adapt our model to Parkinson's disease. Finally, we discuss the choices we made to take into account the temporal aspect of our problem, both during the ground truth creation using the long-term evolution of a cognitive score, and for the choice of using pairs of visits as input instead of longer sequences
Kherroubi, Zine el abidine. "Novel off-board decision-making strategy for connected and autonomous vehicles (Use case highway : on-ramp merging)." Thesis, Lyon, 2020. http://www.theses.fr/2020LYSE1331.
Повний текст джерелаMerging in the highway on-ramp is a significant challenge toward realizing fully automated driving (level 4 of autonomous driving). The combination of communication technology and autonomous driving technology, which underpins the notion of Connected Autonomous Vehicles (CAVs), may improve greatly safety performances when performing highway on-ramp merging. However, even with the emergence of CAVs vehicles, some keys constraints should be considered to achieve a safe on-ramp merging. First, human-driven vehicles will still be present on the road, and it may take decades before all the commercialized vehicles will be fully autonomous and connected. Also, on-board vehicle sensors may provide inaccurate or incomplete data due to sensors limitations and blind spots, especially in such critical situations. To resolve these issues, the present thesis introduces a novel solution that uses an off-board Road-Side Unit (RSU) to realize fully automated highway on-ramp merging for connected and automated vehicles. Our proposed approach is based on an Artificial Neural Network (ANN) to predict drivers’ intentions. This prediction is used as an input state to a Deep Reinforcement Learning (DRL) agent that outputs the longitudinal acceleration for the merging vehicle. To achieve this, we first show how the road-side unit may be used to enhance perception in the on-ramp zone. We then propose a driver intention model that can predict the behavior of the human-driven vehicles in the main highway lane, with 99% accuracy. We use the output of this model as an input state to train a Twin Delayed Deep Deterministic Policy Gradients (TD3) agent that learns « safe » and « cooperative » driving policy to perform highway on-ramp merging. We show that our proposed decision-making strategy improves performance compared to the solutions proposed previously
Lesner, Boris. "Planification et apprentissage par renforcement avec modèles d'actions compacts." Caen, 2011. http://www.theses.fr/2011CAEN2074.
Повний текст джерелаWe study Markovian Decision Processes represented with Probabilistic STRIPS action models. A first part of our work is about solving those processes in a compact way. To that end we propose two algorithms. A first one based on propositional formula manipulation allows to obtain approximate solutions in tractable propositional fragments such as Horn and 2-CNF. The second algorithm solves exactly and efficiently problems represented in PPDDL using a new notion of extended value functions. The second part is about learning such action models. We propose different approaches to solve the problem of ambiguous observations occurring while learning. Firstly, a heuristic method based on Linear Programming gives good results in practice yet without theoretical guarantees. We next describe a learning algorithm in the ``Know What It Knows'' framework. This approach gives strong theoretical guarantees on the quality of the learned models as well on the sample complexity. These two approaches are then put into a Reinforcement Learning setting to allow an empirical evaluation of their respective performances
Cohen-Hadria, Alice. "Estimation de descriptions musicales et sonores par apprentissage profond." Electronic Thesis or Diss., Sorbonne université, 2019. http://www.theses.fr/2019SORUS607.
Повний текст джерелаIn Music Information Retrieval (MIR) and voice processing, the use of machine learning tools has become in the last few years more and more standard. Especially, many state-of-the-art systems now rely on the use of Neural Networks.In this thesis, we propose a wide overview of four different MIR and voice processing tasks, using systems built with neural networks. More precisely, we will use convolutional neural networks, an image designed class neural networks. The first task presented is music structure estimation. For this task, we will show how the choice of input representation can be critical, when using convolutional neural networks. The second task is singing voice detection. We will present how to use a voice detection system to automatically align lyrics and audio tracks.With this alignment mechanism, we have created the largest synchronized audio and speech data set, called DALI. Singing voice separation is the third task. For this task, we will present a data augmentation strategy, a way to significantly increase the size of a training set. Finally, we tackle voice anonymization. We will present an anonymization method that both obfuscate content and mask the speaker identity, while preserving the acoustic scene
Maillard, Odalric-Ambrym. "APPRENTISSAGE SÉQUENTIEL : Bandits, Statistique et Renforcement." Phd thesis, Université des Sciences et Technologie de Lille - Lille I, 2011. http://tel.archives-ouvertes.fr/tel-00845410.
Повний текст джерелаKlein, Édouard. "Contributions à l'apprentissage par renforcement inverse." Thesis, Université de Lorraine, 2013. http://www.theses.fr/2013LORR0185/document.
Повний текст джерелаThis thesis, "Contributions à l'apprentissage par renforcement inverse", brings three major contributions to the community. The first one is a method for estimating the feature expectation, a quantity involved in most of state-of-the-art approaches which were thus extended to a batch off-policy setting. The second major contribution is an Inverse Reinforcement Learning algorithm, structured classification for inverse reinforcement learning (SCIRL), which relaxes a standard constraint in the field, the repeated solving of a Markov Decision Process, by introducing the temporal structure (using the feature expectation) of this process into a structured margin classification algorithm. The afferent theoritical guarantee and the good empirical performance it exhibited allowed it to be presentend in a good international conference: NIPS. Finally, the third contribution is cascaded supervised learning for inverse reinforcement learning (CSI) a method consisting in learning the expert's behavior via a supervised learning approach, and then introducing the temporal structure of the MDP via a regression involving the score function of the classifier. This method presents the same type of theoretical guarantee as SCIRL, but uses standard components for classification and regression, which makes its use simpler. This work will be presented in another good international conference: ECML
Gelly, Sylvain. "Une contribution à l'apprentissage par renforcement : application au Computer Go." Paris 11, 2007. http://www.theses.fr/2007PA112227.
Повний текст джерелаReinforcement Learning (RL) is at the interface of control theory, supervised and unsupervised learning, optimization and cognitive sciences. While RL addresses many objectives with major economic impact, it raises deep theoretical and practical difficulties. This thesis brings some contributions to RL, mainly on three axis. The first axis corresponds to environment modeling, i. E. Learning the transition function between two time steps. Factored approaches give an efficiently framework for the learning and use of this model. The Bayesian Networks are a tool to represent such a model, and this work brings new learning criterion, either in parametric learning (conditional probabilities) and non parametric (structure). The second axis is a study in continuous space and action RL, thanks to the dynamic programming algorithm. This analysis tackles three fundamental steps: optimization (action choice from the value function), supervised learning (regression) of the value function and choice of the learning examples (active learning). The third axis tackles the applicative domain of the game of Go, as a high dimensional discrete control problem, one of the greatest challenge in Machine Learning. The presented algorithms with their improvements made the resulting program, MoGo, win numerous international competitions, becoming for example the first go program playing at an amateur dan level on 9x9
Zaidenberg, Sofia. "Apprentissage par renforcement de modèles de contexte pour l'informatique ambiante." Grenoble INPG, 2009. http://www.theses.fr/2009INPG0088.
Повний текст джерелаThis thesis studies the automatic acquisition by machine learning of a context model for a user in a ubiquitous environment. In such an environment, devices can communicate and cooperate in order to create a consistent computerized space. Some devices possess perceptual capabilities. The environment uses them to detect the user's situation his context. Other devices are able to execute actions. Our problematics consists in determining the optimal associations, for a given user, between situations and actions. Machine learning seems to be a sound approach since it results in a customized environment without requiring an explicit specification from the user. A life long learning lets the environment adapt itself continuously to world changes and user preferences changes. Reinforcement learning can be a solution to this problem, as long as it is adapted to some particular constraints due to our application setting
Degris, Thomas. "Apprentissage par renforcement dans les processus de décision Markoviens factorisés." Paris 6, 2007. http://www.theses.fr/2007PA066594.
Повний текст джерелаMoukari, Michel. "Estimation de profondeur à partir d'images monoculaires par apprentissage profond." Thesis, Normandie, 2019. http://www.theses.fr/2019NORMC211/document.
Повний текст джерелаComputer vision is a branch of artificial intelligence whose purpose is to enable a machine to analyze, process and understand the content of digital images. Scene understanding in particular is a major issue in computer vision. It goes through a semantic and structural characterization of the image, on one hand to describe its content and, on the other hand, to understand its geometry. However, while the real space is three-dimensional, the image representing it is two-dimensional. Part of the 3D information is thus lost during the process of image formation and it is therefore non trivial to describe the geometry of a scene from 2D images of it.There are several ways to retrieve the depth information lost in the image. In this thesis we are interested in estimating a depth map given a single image of the scene. In this case, the depth information corresponds, for each pixel, to the distance between the camera and the object represented in this pixel. The automatic estimation of a distance map of the scene from an image is indeed a critical algorithmic brick in a very large number of domains, in particular that of autonomous vehicles (obstacle detection, navigation aids).Although the problem of estimating depth from a single image is a difficult and inherently ill-posed problem, we know that humans can appreciate distances with one eye. This capacity is not innate but acquired and made possible mostly thanks to the identification of indices reflecting the prior knowledge of the surrounding objects. Moreover, we know that learning algorithms can extract these clues directly from images. We are particularly interested in statistical learning methods based on deep neural networks that have recently led to major breakthroughs in many fields and we are studying the case of the monocular depth estimation
Bertrand, Hadrien. "Optimisation d'hyper-paramètres en apprentissage profond et apprentissage par transfert : applications en imagerie médicale." Thesis, Université Paris-Saclay (ComUE), 2019. http://www.theses.fr/2019SACLT001/document.
Повний текст джерелаIn the last few years, deep learning has changed irrevocably the field of computer vision. Faster, giving better results, and requiring a lower degree of expertise to use than traditional computer vision methods, deep learning has become ubiquitous in every imaging application. This includes medical imaging applications. At the beginning of this thesis, there was still a strong lack of tools and understanding of how to build efficient neural networks for specific tasks. Thus this thesis first focused on the topic of hyper-parameter optimization for deep neural networks, i.e. methods for automatically finding efficient neural networks on specific tasks. The thesis includes a comparison of different methods, a performance improvement of one of these methods, Bayesian optimization, and the proposal of a new method of hyper-parameter optimization by combining two existing methods: Bayesian optimization and Hyperband.From there, we used these methods for medical imaging applications such as the classification of field-of-view in MRI, and the segmentation of the kidney in 3D ultrasound images across two populations of patients. This last task required the development of a new transfer learning method based on the modification of the source network by adding new geometric and intensity transformation layers.Finally this thesis loops back to older computer vision methods, and we propose a new segmentation algorithm combining template deformation and deep learning. We show how to use a neural network to predict global and local transformations without requiring the ground-truth of these transformations. The method is validated on the task of kidney segmentation in 3D US images
Darwiche, Domingues Omar. "Exploration en apprentissage par renforcement : au-delà des espaces d'états finis." Thesis, Université de Lille (2022-....), 2022. http://www.theses.fr/2022ULILB002.
Повний текст джерелаReinforcement learning (RL) is a powerful machine learning framework to design algorithms that learn to make decisions and to interact with the world. Algorithms for RL can be classified as offline or online. In the offline case, the algorithm is given a fixed dataset, based on which it needs to compute a good decision-making strategy. In the online case, an agent needs to efficiently collect data by itself, by interacting with the environment: that is the problem of exploration in reinforcement learning. This thesis presents theoretical and practical contributions to online RL. We investigate the worst-case performance of online RL algorithms in finite environments, that is, those that can be modeled with a finite amount of states, and where the set of actions that can be taken by an agent is also finite. Such performance degrades as the number of states increases, whereas in real-world applications the state set can be arbitrarily large or continuous. To tackle this issue, we propose kernel-based algorithms for exploration that can be implemented for general state spaces, and for which we provide theoretical results under weak assumptions on the environment. Those algorithms rely on a kernel function that measures the similarity between different states, which can be defined on arbitrary state-spaces, including discrete sets and Euclidean spaces, for instance. Additionally, we show that our kernel-based algorithms are able to handle non-stationary environments by using time-dependent kernel functions, and we propose and analyze approximate versions of our methods to reduce their computational complexity. Finally, we introduce a scalable approximation of our kernel-based methods, that can be implemented with deep reinforcement learning and integrate different representation learning methods to define a kernel function
Garcia, Pascal. "Exploration guidée et induction de comportements génériques en apprentissage par renforcement." Rennes, INSA, 2004. http://www.theses.fr/2004ISAR0010.
Повний текст джерелаReinforcement learning is a general framework in which an autonomous agent learns which actions to choose in particular situations (states) in order to optimize some reinforcements (rewards or punitions) in the long run. Even if a lot of tasks can be formulated in this framework, there are two problems with the standard reinforcement learning algorithms: 1. Due to the learning time of those algorithms, in practice, tasks with a moderatly large state space are not solvable in reasonable time. 2. Given several problems to solve in some domains, a standard reinforcement learning agent learns an optimal policy from scratch for each problem. It would be far more useful to have systems that can solve several problems over time, using the knowledge obtained from previous problem instances to guide in learning on new problems. We propose some methods to address those issues: 1. We define two formalisms to introduce a priori knowledge to guide the agent on a given task. The agent has an initial behaviour which can be modified during the learning process. 2. We define a method to induce generic behaviours,based on the previously solved tasks and on basicbuilding blocks. Those behaviours will be added to the primitive actions of a new related task tohelp the agent solve it
Routhier, Etienne. "Conception de séquences génomiques artificielles chez la levure par apprentissage profond." Thesis, Sorbonne université, 2021. http://www.theses.fr/2021SORUS465.
Повний текст джерелаRecent technological advances in the field of biotechnologies such as CRISPR and the de novo DNA oligonucleotides synthesis now make it possible to modify precisely and intensively genomes. Projects aiming to design partially or completely synthetic genomes, in particular yeast genomes, have been developed by taking advantage of these technologies. However, to achieve this goal it is necessary to control the activity of artificial sequences, which remains a challenge today. Fortunately, the recent emergence of deep learning methodologies able to recognize the genomic function associated to a DNA sequence seems to provide a powerful tool for anticipating the activity of synthetic genomes and facilitating their design. In this perspective, we propose to use deep learning methodologies in order to design synthetic yeast sequences controlling the local structure of the genome. In particular, I will present the methodology we have developed in order to design synthetic sequences precisely positioning nucleosomes - a molecule determining the structure of DNA at the lowest scale - in yeast. I will also show that this methodology opens up the prospect of designing sequences controlling the immediately higher level of structure: loops. The design of sequences controlling the local structure makes it possible to precisely identify the determinants of this structure
Liu, Zongyi. "Self-Adaptive Bandwidth Control for Balanced QoS and Energy Aware Optimization in Wireless Sensor Network." Thesis, Toulouse, INSA, 2017. http://www.theses.fr/2017ISAT0034/document.
Повний текст джерелаIn the Wireless Multimedia Sensor Networks (WMSNs) field, highly saturated flow increases the probability of collision and congestion in data transmission which dramatically degrade the performance of Quality of Service (QoS). Multi-channels deployment technique is often applied to parallel transmission for QoS guarantee. However, how to make trade-off between QoS requirement and energy efficiency is a challenges to energy-constrained WMSNs. Theoretical analysis of MAC layer and PHY layer structure based on IEEE 802.15.4 standard, aim to study on the cross-layer analytical model in order to provide stronger understanding on the relationship between sensor network parameters and performance, pave the way for new enhancements in succedent multi-channel optimization research. Find effective performance indicator and design efficient performance collection or estimation approach based on the corresponding metrics, which could be used as the parameter input of multi-channel assignment mechanism. Comprehensive dynamically control system is designed for multi-channel assignment task based on light weight and high efficient computation intelligence techniques. We present a fuzzy-based dynamic bandwidth multi-channel assignment mechanism (MCDB_FLS). Cross-layer proactive available bandwidth is estimated as parameters for multi-channel deployment admission control. Reinforcement learning-based approach is proposed for more wisely decision-making in multi- channel allocation mission. Furthermore, fuzzy logic-based bandwidth threshold model provides dynamic optimization on system admission control. Simulations show the MCDB_FLS performs better than benchmark on the metrics of QoS and energy efficiency, achieves the trade-off between energy efficiency and QoS improvement. Finally, we introduce the integration of incremental machine learning approach into multi-channel assignment mechanism with Deep Q Network reinforcement learning method (DQMC). Besides, fully action weight initialization is implemented based on multi-class supervised learning classifier with stacking ensemble approach. DQMC improve the ability of self-adaptive and smart control to learn pattern from different environment of multi-tasks WMSNs
Zhang, Ping. "Etudes de différents aspects de l'apprentissage par renforcement." Compiègne, 1997. http://www.theses.fr/1997COMP0993.
Повний текст джерелаThis dissertation deals with the research on three important aspects of the reinforcement learning : the temporal differences (TD(). ), the Q-learning and the exploration/ exploitation dilemma. We propose algorithms and techniques based on new concepts that allow a better understanding, and ultimately, the solution to the problem of reinforcement learning. The first part of this work deals with a method that optimizes the choice of parameter of T D(). . ) and then solves a real problem of a person's ability to evaluate utilizing the different methods based on the principle of T D(>,). In the second part, we introduce the notion "confidence" and propose a new version of Q-learning, SCIQ, which generalizes and improves the Q-learning. We point out that this algorithm can overcome the over-estimation problem of Q-values associated with non-optimal actions. Contrary to other versions of Q-learning, our algorithm is adaptive thanks to its evolving capacity to modify the Q-values. Again, it is robust and faster than the Q-learning. In the last part, in order to solve the exploration/exploitation dilemma, the notion "entropy" is introduced as the measure of information on the system state. We present two methods allowing to estimate the entropy approximation and two types of tech¬niques for exploration by means of these estimations. It is noted that aside from using entropy itself by using the entropy approximation we can define the efficient algorithm without the counter and extra structure
Poiron-Guidoni, Nicolas. "Apports des méthodes d’optimisation et du calcul haute performance à la théorie de la modélisation et de la simulation : application à la gestion des ressources halieutiques." Thesis, Corte, 2021. http://www.theses.fr/2021CORT0013.
Повний текст джерелаThe computer science project (SiSU) of the CNRS Science for the Environment Joint Research Unit designs decision support methods to help better management of complex environmental systems.This thesis work is part of this context. They aim to study the contributions of several types of computer methods to improve our knowledge of complex systems and thus provide assistance in their management in situations of high uncertainty. Indeed, complex environmental systems cannot always be known and modeled with precision. This is for example the case in fisheries biology where management methods must be proposed despite a lack of knowledge on the observed system, in our case study: the Corsican coastal fishery. Our first work focused on the calibration of models, i.e. the search for parameter values allowing our models to best represent the dynamics of the system. They have shown the limits of the usual approaches and the need to use probabilistic approaches based on large quantities of simulations. They bring a precious help for the acquisition of knowledge, in particular by delimiting sets of solutions. These sets can then be used in robust optimization methods, or even in adjustable robust optimization. These approaches allow not only to take into account the uncertainties, but also to quantify the reduction of uncertainty that new years of data can bring, in order to propose more and more precise strategies in the long term. Optimization can therefore be used effectively at the level of decision makers. However, the small-scale coastal fishery in Corsica is a system in which a large number of actors act with different behaviors that are difficult to predict and control. Optimization does not seem adapted to the study of this scale because of the quantity of parameters and the infinite number of stochastic transitions generated. For this, methods based on deep reinforcement learning have been proposed. These approaches allowed us to propose a model that manages both decision-makers and fishermen, the former seeking to reduce the ecological impact, the latter to maximize their gains. From this, we were able to show that little knowledge is sufficient for the maximization of the fishermen's gains. Moreover, this approach, coupled with optimization, allowed us to obtain efficient quota decisions. Finally, this system allowed us to study the impact of certain individual behaviors of maximizing gains to the detriment of respecting the recommendations of the decision makers. It then appeared that effective and adapted management policies can help to mitigate the ecological impact of a significant amount of these behaviors. Thus, we were able to contribute in a theoretical way to broaden the application domains of the theory of modeling and simulation, to propose a set of optimization and machine learning tools for the management of dynamic systems partially observable, but also applicative for the problem of fisheries management in Corsica
Dahmani, Sara. "Synthèse audiovisuelle de la parole expressive : modélisation des émotions par apprentissage profond." Electronic Thesis or Diss., Université de Lorraine, 2020. http://www.theses.fr/2020LORR0137.
Повний текст джерела: The work of this thesis concerns the modeling of emotions for expressive audiovisual textto-speech synthesis. Today, the results of text-to-speech synthesis systems are of good quality, however audiovisual synthesis remains an open issue and expressive synthesis is even less studied. As part of this thesis, we present an emotions modeling method which is malleable and flexible, and allows us to mix emotions as we mix shades on a palette of colors. In the first part, we present and study two expressive corpora that we have built. The recording strategy and the expressive content of these corpora are analyzed to validate their use for the purpose of audiovisual speech synthesis. In the second part, we present two neural architectures for speech synthesis. We used these two architectures to model three aspects of speech : 1) the duration of sounds, 2) the acoustic modality and 3) the visual modality. First, we use a fully connected architecture. This architecture allowed us to study the behavior of neural networks when dealing with different contextual and linguistic descriptors. We were also able to analyze, with objective measures, the network’s ability to model emotions. The second neural architecture proposed is a variational auto-encoder. This architecture is able to learn a latent representation of emotions without using emotion labels. After analyzing the latent space of emotions, we presented a procedure for structuring it in order to move from a discrete representation of emotions to a continuous one. We were able to validate, through perceptual experiments, the ability of our system to generate emotions, nuances of emotions and mixtures of emotions, and this for expressive audiovisual text-to-speech synthesis
Bhattarai, Binod. "Développement de méthodes de rapprochement physionomique par apprentissage machine." Caen, 2016. https://hal.archives-ouvertes.fr/tel-01467985.
Повний текст джерелаThe work presented in this PhD thesis takes place in the general context of face matching. More precisely, our goal is to design and develop novel algorithms to learn compact, discriminative, domain invariant or de-identifying representations of faces. Searching and indexing faces open the door to many interesting applications. However, this is made day after day more challenging due to the rapid growth of the volume of faces to analyse. Representing faces by compact and discriminative features is consequently es- sential to deal with such very large datasets. Moreover, this volume is increasing without any apparent limits; this is why it is also relevant to propose solutions to organise faces in meaningful ways, in order to reduce the search space and improve efficiency of the retrieval. Although the volume of faces available on the internet is increasing, it is still difficult to find annotated examples to train models for each possible use cases e. G. For different races, sexes, etc. For every specifie task. Learning a model with training examples from a group of people can fail to predict well in another group due to the uneven rate of changes of biometrie dimensions e. G. , ageing, among them. Similarly, a modellean1ed from a type of feature can fail to make good predictions when tested with another type of feature. It would be ideal to have models producing face representations that would be invariant to these discrepancies. Learning common representations ultimately helps to reduce the domain specifie parameters and, more important!y, allows to use training examples from domains weil represented to other demains. Hence, there is a need for designing algorithms to map the features from different domains to a common subspace -bringing faces bearing same properties closer. On the other band, as automatic face matching tools are getting smarter and smarter, there is an increasing threat on privacy. The popularity in photo sharing on the social networks has exacerbated this risk. In such a context, altering the representations of faces so that the faces cannot be identified by automatic face matchers -while the faces look as similar as before -has become an interesting perspective toward privacy protection. It allows users to limit the risk of sharing their photos in social networks. In ali these scenarios, we explored how the use of Metric Leaming methods as weil as those of Deep Learning can help us to leam compact and discriminative representations of faces. We build on these tools, proposing compact, discriminative, domain invariant representations and de-identifying representations of faces crawled from Flicker. Corn to LFW and generated a novel and more challenging dataset to evaluate our algorithms in large-scale. We applied the proposed methods on a wide range of facial analysing applications. These applications include: large-scale face retrieval, age estimation, attribute predictions and identity de-identification. We have evaluated our algorithms on standard and challenging public datasets such as: LFW, CelebA, MORPH II etc. Moreover, we appended lM faces crawled from Flicker. Corn to LFW and generated a novel and more challenging dataset to evaluate our algorithms in large-scale. Our experiments show that the proposed methods are more accurate and more efficient than compared competitive baselines and existing state-of-art methods, and attain new state-of-art performance
Laurent, Guillaume. "Synthèse de comportements par apprentissages par renforcement parallèles : application à la commande d'un micromanipulateur plan." Phd thesis, Université de Franche-Comté, 2002. http://tel.archives-ouvertes.fr/tel-00008761.
Повний текст джерелаPhilip, Julien. "Édition et rendu à base d’images multi-vues par apprentissage profond et optimisation." Thesis, Université Côte d'Azur, 2020. http://www.theses.fr/2020COAZ4048.
Повний текст джерелаComputer-generated imagery (CGI) takes a growing place in our everyday environment. Whether it is in video games or movies, CGI techniques are constantly improving in quality but also require ever more qualitative artistic content which takes a growing time to create. With the emergence of virtual and augmented reality, often comes the need to render or re-render assets that exist in our world. To allow widespread use of CGI in applications such as telepresence or virtual visits, the need for manual artistic replication of assets must be removed from the process. This can be done with the help of Image-Based Rendering (IBR) techniques that allow scenes or objects to be rendered in a free-viewpoint manner from a set of sparse input photographs. While this process requires little to no artistic work, it also does not allow for artistic control or editing of scene content. In this dissertation, we explore Multi-view Image Editing and Rendering. To allow casually captured scenes to be rendered with content alterations such as object removal, lighting edition, or scene compositing, we leverage the use of optimization techniques and modern deep-learning. We design our methods to take advantage of all the information present in multi-view content while handling specific constraints such as multi-view coherency. For object removal, we introduce a new plane-based multi-view inpainting algorithm. Planes are a simple yet effective way to fill geometry and they naturally enforce multi-view coherency as inpainting is computed in a shared rectified texture space, allowing us to correctly respect perspective. We demonstrate instance-based object removal at the scale of a street in scenes composed of several hundreds of images. We next address outdoor relighting with a learning-based algorithm that efficiently allows the illumination in a scene to be changed, while removing and synthesizing cast shadows for any given sun position and accounting for global illumination. An approximate geometric proxy built using multi-view stereo is used to generate illumination and shadow related image buffers that guide a neural network. We train this network on a set of synthetic scenes allowing full supervision of the learning pipeline. Careful data augmentation allows our network to transfer to real scenes and provides state of the art relighting results. We also demonstrate the capacity of this network to be used to compose real scenes captured under different lighting conditions and orientation. We then present contributions to image-based rendering quality. We discuss how our carefully designed depth-map meshing and simplification algorithm improve rendering performance and quality of a new learning-based IBR method. Finally, we present a method that combines relighting, IBR, and material analysis. To enable relightable IBR with accurate glossy effects, we extract both material appearance variations and qualitative texture information from multi-view content in the form of several IBR heuristics. We further combine them with path-traced irradiance images that specify the input and target lighting. This combination allows a neural network to be trained to implicitly extract material properties and produce realistic-looking relit viewpoints. Separating diffuse and specular supervision is crucial in obtaining high-quality output
Bouzid, Salah Eddine. "Optimisation multicritères des performances de réseau d’objets communicants par méta-heuristiques hybrides et apprentissage par renforcement." Thesis, Le Mans, 2020. http://cyberdoc-int.univ-lemans.fr/Theses/2020/2020LEMA1026.pdf.
Повний текст джерелаThe deployment of Communicating Things Networks (CTNs), with continuously increasing densities, needs to be optimal in terms of quality of service, energy consumption and lifetime. Determining the optimal placement of the nodes of these networks, relative to the different quality criteria, is an NP-Hard problem. Faced to this NP-Hardness, especially for indoor environments, existing approaches focus on the optimization of one single objective while neglecting the other criteria, or adopt an expensive manual solution. Finding new approaches to solve this problem is required. Accordingly, in this thesis, we propose a new approach which automatically generates the deployment that guarantees optimality in terms of performance and robustness related to possible topological failures and instabilities. The proposed approach is based, on the first hand, on the modeling of the deployment problem as a multi-objective optimization problem under constraints, and its resolution using a hybrid algorithm combining genetic multi-objective optimization with weighted sum optimization and on the other hand, the integration of reinforcement learning to guarantee the optimization of energy consumption and the extending the network lifetime. To apply this approach, two tools are developed. A first called MOONGA (Multi-Objective Optimization of wireless Network approach based on Genetic Algorithm) which automatically generates the placement of nodes while optimizing the metrics that define the QoS of the CTN: connectivity, m-connectivity, coverage, k-coverage, coverage redundancy and cost. MOONGA tool considers constraints related to the architecture of the deployment space, the network topology, the specifies of the application and the preferences of the network designer. The second optimization tool is named R2LTO (Reinforcement Learning for Life-Time Optimization), which is a new routing protocol for CTNs, based on distributed reinforcement learning that allows to determine the optimal rooting path in order to guarantee energy-efficiency and to extend the network lifetime while maintaining the required QoS
Buffet, Olivier. "Une double approche modulaire de l'apprentissage par renforcement pour des agents intelligents adaptatifs." Phd thesis, Université Henri Poincaré - Nancy I, 2003. http://tel.archives-ouvertes.fr/tel-00509349.
Повний текст джерелаDutech, Alain. "Apprentissage par Renforcement : Au delà des Processus Décisionnels de Markov (Vers la cognition incarnée)." Habilitation à diriger des recherches, Université Nancy II, 2010. http://tel.archives-ouvertes.fr/tel-00549108.
Повний текст джерелаCoulom, Rémi. "Apprentissage par renforcement utilisant des réseaux de neurones avec des applications au contrôle moteur." Phd thesis, Grenoble INPG, 2002. http://tel.archives-ouvertes.fr/tel-00004386.
Повний текст джерелаZhang, Jian. "Modèles de Mobilité de Véhicules par Apprentissage Profond dans les Systèmes de Tranport Intelligents." Thesis, Ecole centrale de Lille, 2018. http://www.theses.fr/2018ECLI0015/document.
Повний текст джерелаThe intelligent transportation systems gain great research interests in recent years. Although the realistic traffic simulation plays an important role, it has not received enough attention. This thesis is devoted to studying the traffic simulation in microscopic level, and proposes corresponding vehicular mobility models. Using deep learning methods, these mobility models have been proven with a promising credibility to represent the vehicles in real-world. Firstly, a data-driven neural network based mobility model is proposed. This model comes from real-world trajectory data and allows mimicking local vehicle behaviors. By analyzing the performance of this basic learning based mobility model, we indicate that an improvement is possible and we propose its specification. An HMM is then introduced. The preparation of this integration is necessary, which includes an examination of traditional dynamics based mobility models and the adaptation method of “classical” models to our situation. At last, the enhanced model is presented, and a sophisticated scenario simulation is built with it to validate the theoretical results. The performance of our mobility model is promising and implementation issues have also been discussed
Gueguen, Maëlle. "Dynamique intracérébrale de l'apprentissage par renforcement chez l'humain." Thesis, Université Grenoble Alpes (ComUE), 2017. http://www.theses.fr/2017GREAS042/document.
Повний текст джерелаWe make decisions every waking day of our life. Facing our options, we tend to pick the most likely to get our expected outcome. Taking into account our past experiences and their outcome is mandatory to identify the best option. This cognitive process is called reinforcement learning. To date, the underlying neural mechanisms are debated. Despite a consensus on the role of dopaminergic neurons in reward processing, several hypotheses on the neural bases of reinforcement learning coexist: either two distinct opposite systems covering cortical and subcortical areas, or a segregation of neurons within brain regions to process reward-based and punishment-avoidance learning.This PhD work aimed to identify the brain dynamics of human reinforcement learning. To unravel the neural mechanisms involved, we used intracerebral recordings in refractory epileptic patients during a probabilistic learning task. In the first study, we used a computational model to tackle the brain dynamics of reinforcement signal encoding, especially the encoding of reward and punishment prediction errors. Local field potentials exhibited the central role of high frequency gamma activity (50-150Hz) in these encodings. We report a role of the ventromedial prefrontal cortex in reward prediction error encoding while the anterior insula and the dorsolateral prefrontal cortex encoded punishment prediction errors. In addition, the magnitude of the neural response in the insula predicted behavioral learning and trial-to-trial behavioral adaptations. These results are consistent with the existence of two distinct opposite cortical systems processing reward and punishments during reinforcement learning. In a second study, we recorded the neural activity of the anterior and dorsomedial nuclei of the thalamus during the same cognitive task. Local field potentials recordings highlighted the role of low frequency theta activity in punishment processing, supporting an implication of these nuclei during punishment-avoidance learning. In a third behavioral study, we investigated the influence of risk on reinforcement learning. We observed a risk-aversion during punishment-avoidance, affecting the performance, as well as a risk-seeking behavior during reward-seeking, revealed by an increased reaction time towards appetitive risky choices. Taken together, these results suggest we are risk-seeking when we have something to gain and risk-averse when we have something to lose, in contrast to the prediction of the prospect theory.Improving our common knowledge of the brain dynamics of human reinforcement learning could improve the understanding of cognitive deficits of neurological patients, but also the decision bias all human beings can exhibit
Godbout, Mathieu. "Approches par bandit pour la génération automatique de résumés de textes." Master's thesis, Université Laval, 2021. http://hdl.handle.net/20.500.11794/69488.
Повний текст джерелаThis thesis discusses the use of bandit methods to solve the problem of training extractive abstract generation models. The extractive models, which build summaries by selecting sentences from an original document, are difficult to train because the target summary of a document is usually not built in an extractive way. It is for this purpose that we propose to see the production of extractive summaries as different bandit problems, for which there exist algorithms that can be leveraged for training summarization models.In this paper, BanditSum is first presented, an approach drawn from the literature that sees the generation of the summaries of a set of documents as a contextual bandit problem. Next,we introduce CombiSum, a new algorithm which formulates the generation of the summary of a single document as a combinatorial bandit. By exploiting the combinatorial formulation,CombiSum manages to incorporate the notion of the extractive potential of each sentence of a document in its training. Finally, we propose LinCombiSum, the linear variant of Com-biSum which exploits the similarities between sentences in a document and uses the linear combinatorial bandit formulation instead
Montagne, Fabien. "Une architecture logicielle pour aider un agent apprenant par renforcement." Littoral, 2008. http://www.theses.fr/2008DUNK0198.
Повний текст джерелаThis thesis deals with reinforcement learning. One of the main advantage of this learning is to not require to know explicitely the expected behavior. During its learning, the agent percieves states, gets a set of rewards and selects actions to carry out. The agent fits its behavior by optimizing the amount of rewards. Nevertheless, the computing time required quickly becomes prohibitive. This is mainly due to the agent’s need of exploring its environment. The approach considered here consists in using external knowledge to “guide” the agent during its exploration. This knowledge constitutes an help which can, for example, be expressed by trajectories that set up a knowledge database. These trajectories are used to limit the exploration of the environment while allowing the agent to build a good quality behavior. Helping an agent does neither involve knowing the actions choose in all states, nor having the same perceptions as the agent. The critic-critic architecture was devised to fulfill to this problematic. It combines a standard reinforcement learning algorithm with an help given through potentials. The potentials assiociate a value to each transition of the trajectories. The value function estimation by the agent and the potential of the help are combined during the training. Fitting this combine dynamically makes it possible to throw assistance into question while guaranteing an optimal or almost optimal policy quickly. It is formally proved that the proposed algorithm converges under certain conditions. Moreover, empirical work show that the agent is able to benefit from an help without these conditions
Geist, Matthieu. "Optimisation des chaînes de production dans l'industrie sidérurgique : une approche statistique de l'apprentissage par renforcement." Phd thesis, Université de Metz, 2009. http://tel.archives-ouvertes.fr/tel-00441557.
Повний текст джерела