Rozprawy doktorskie na temat „Réseaux de neurones récurrents convolutifs”
Utwórz poprawne odniesienie w stylach APA, MLA, Chicago, Harvard i wielu innych
Sprawdź 50 najlepszych rozpraw doktorskich naukowych na temat „Réseaux de neurones récurrents convolutifs”.
Przycisk „Dodaj do bibliografii” jest dostępny obok każdej pracy w bibliografii. Użyj go – a my automatycznie utworzymy odniesienie bibliograficzne do wybranej pracy w stylu cytowania, którego potrzebujesz: APA, MLA, Harvard, Chicago, Vancouver itp.
Możesz również pobrać pełny tekst publikacji naukowej w formacie „.pdf” i przeczytać adnotację do pracy online, jeśli odpowiednie parametry są dostępne w metadanych.
Przeglądaj rozprawy doktorskie z różnych dziedzin i twórz odpowiednie bibliografie.
Shahkarami, Abtin. "Complexity reduction over bi-RNN-based Kerr nonlinearity equalization in dual-polarization fiber-optic communications via a CRNN-based approach". Electronic Thesis or Diss., Institut polytechnique de Paris, 2022. http://www.theses.fr/2022IPPAT034.
Pełny tekst źródłaThe impairments arising from the Kerr nonlinearity in optical fibers limit the achievable information rates in fiber-optic communication. Unlike linear effects, such as chromatic dispersion and polarization-mode dispersion, which can be compensated via relatively simple linear equalization at the receiver, the computational complexity of the conventional nonlinearity mitigation techniques, such as the digital backpropagation, can be substantial. Neural networks have recently attracted attention, in this context, for low-complexity nonlinearity mitigation in fiber-optic communications. This Ph.D. dissertation deals with investigating the recurrent neural networks to efficiently compensate for the nonlinear channel impairments in dual-polarization long-haul fiber-optic transmission. We present a hybrid convolutional recurrent neural network (CRNN) architecture, comprising a convolutional neural network (CNN) -based encoder followed by a recurrent layer working in tandem. The CNN-based encoder represents the shortterm channel memory arising from the chromatic dispersion efficiently, while transitioning the signal to a latent space with fewer relevant features. The subsequent recurrent layer is implemented in the form of a unidirectional vanilla RNN, responsible for capturing the long-range interactions neglected by the CNN encoder. We demonstrate that the proposed CRNN achieves the performance of the state-of-theart equalizers in optical fiber communication, with significantly lower computational complexity depending on the system model. Finally, the performance complexity trade-off is established for a number of models, including multi-layer fully-connected neural networks, CNNs, bidirectional recurrent neural networks, bidirectional long short-term memory (bi-LSTM), bidirectional gated recurrent units, convolutional bi-LSTM models, and the suggested hybrid model
Barhoumi, Amira. "Une approche neuronale pour l’analyse d’opinions en arabe". Thesis, Le Mans, 2020. http://www.theses.fr/2020LEMA1022.
Pełny tekst źródłaMy thesis is part of Arabic sentiment analysis. Its aim is to determine the global polarity of a given textual statement written in MSA or dialectal arabic. This research area has been subject of numerous studies dealing with Indo-European languages, in particular English. One of difficulties confronting this thesis is the processing of Arabic. In fact, Arabic is a morphologically rich language which implies a greater sparsity : we want to overcome this problem by producing, in a completely automatic way, new arabic specific embeddings. Our study focuses on the use of a neural approach to improve polarity detection, using embeddings. These embeddings have revealed fundamental in various natural languages processing tasks (NLP). Our contribution in this thesis concerns several axis. First, we begin with a preliminary study of the various existing pre-trained word embeddings resources in arabic. These embeddings consider words as space separated units in order to capture semantic and syntactic similarities in the embedding space. Second, we focus on the specifity of Arabic language. We propose arabic specific embeddings that take into account agglutination and morphological richness of Arabic. These specific embeddings have been used, alone and in combined way, as input to neural networks providing an improvement in terms of classification performance. Finally, we evaluate embeddings with intrinsic and extrinsic methods specific to sentiment analysis task. For intrinsic embeddings evaluation, we propose a new protocol introducing the notion of sentiment stability in the embeddings space. We propose also a qualitaive extrinsic analysis of our embeddings by using visualisation methods
Boutin, Victor. "Etude d’un algorithme hiérarchique de codage épars et prédictif : vers un modèle bio-inspiré de la perception visuelle". Thesis, Aix-Marseille, 2020. http://www.theses.fr/2020AIXM0028.
Pełny tekst źródłaBuilding models to efficiently represent images is a central and difficult problem in the machine learning community. The neuroscientific study of the early visual cortical areas is a great source of inspiration to find economical and robust solutions. For instance, Sparse Coding (SC) is one of the most successful frameworks to model neural computation at the local scale in the visual cortex. At the structural scale of the ventral visual pathways, the Predictive Coding (PC) theory has been proposed to model top-down and bottom-up interaction between cortical regions. The presented thesis introduces a model called the Sparse Deep Predictive Coding (SDPC) that combines Sparse Coding and Predictive Coding in a hierarchical and convolutional architecture. We analyze the SPDC from a computational and a biological perspective. In terms of computation, the recurrent connectivity introduced by the PC framework allows the SDPC to converge to lower prediction errors with a higher convergence rate. In addition, we combine neuroscientific evidence with machine learning methods to analyze the impact of recurrent processing at both the neural organization and representational level. At the neural organization level, the feedback signal of the model accounted for a reorganization of the V1 association fields that promotes contour integration. At the representational level, the SDPC exhibited significant denoising ability which is highly correlated with the strength of the feedback from V2 to V1. These results from the SDPC model demonstrate that neuro-inspiration might be the right methodology to design more powerful and more robust computer vision algorithms
Pothier, Dominique. "Réseaux convolutifs à politiques". Master's thesis, Université Laval, 2021. http://hdl.handle.net/20.500.11794/69184.
Pełny tekst źródłaDespite their excellent performances, artificial neural networks high demand of both data and computational power limit their adoption in many domains. Developing less demanding architecture thus remain an important endeavor. This thesis seeks to produce a more flexible and less resource-intensive architecture by using reinforcement learning theory. When considering a network as an agent instead of a function approximator, one realize that the implicit policy followed by popular feed forward networks is extremely simple. We hypothesize that an architecture able to learn a more flexible policy could reach similar performances while reducing its resource footprint. The architecture we propose is inspired by research done in weight prediction, particularly by the hypernetwork architecture, which we use as a baseline model.Our results show that learning a dynamic policy achieving similar results to the static policies of conventional networks is not a trivial task. Our proposed architecture succeeds in limiting its parameter space by 20%, but does so at the cost of a 24% computation increase and loss of5% accuracy. Despite those results, we believe that this architecture provides a baseline that can be improved in multiple ways that we describe in the conclusion.
Al, Hajj Hassan. "Video analysis for augmented cataract surgery". Thesis, Brest, 2018. http://www.theses.fr/2018BRES0041/document.
Pełny tekst źródłaThe digital era is increasingly changing the world due to the sheer volume of data produced every day. The medical domain is highly affected by this revolution, because analysing this data can be a source of education/support for the clinicians. In this thesis, we propose to reuse the surgery videos recorded in the operating rooms for computer-assisted surgery system. We are chiefly interested in recognizing the surgical gesture being performed at each instant in order to provide relevant information. To achieve this goal, this thesis addresses the surgical tool recognition problem, with applications in cataract surgery. The main objective of this thesis is to address the surgical tool recognition problem in cataract surgery videos.In the surgical field, those tools are partially visible in videos and highly similar to one another. To address the visual challenges in the cataract surgical field, we propose to add an additional camera filming the surgical tray. Our goal is to detect the tool presence in the two complementary types of videos: tool-tissue interaction and surgical tray videos. The former records the patient's eye and the latter records the surgical tray activities.Two tasks are proposed to perform the task on the surgical tray videos: tools change detection and tool presence detection.First, we establish a similar pipeline for both tasks. It is based on standard classification methods on top of visual learning features. It yields satisfactory results for the tools change task, howev-lateer, it badly performs the surgical tool presence task on the tray. Second, we design deep learning architectures for the surgical tool detection on both video types in order to address the difficulties in manually designing the visual features.To alleviate the inherent challenges on the surgical tray videos, we propose to generate simulated surgical tray scenes along with a patch-based convolutional neural network (CNN).Ultimately, we study the temporal information using RNN processing the CNN results. Contrary to our primary hypothesis, the experimental results show deficient results for surgical tool presence on the tray but very good results on the tool-tissue interaction videos. We achieve even better results in the surgical field after fusing the tool change information coming from the tray and tool presence signals on the tool-tissue interaction videos
Boné, Romuald. "Réseaux de neurones récurrents pour la prévision de séries temporelles". Tours, 2000. http://www.theses.fr/2000TOUR4003.
Pełny tekst źródłaStrock, Anthony. "Mémoire de travail dans les réseaux de neurones récurrents aléatoires". Thesis, Bordeaux, 2020. http://www.theses.fr/2020BORD0195.
Pełny tekst źródłaWorking memory can be defined as the ability to temporarily store and manipulate information of any kind.For example, imagine that you are asked to mentally add a series of numbers.In order to accomplish this task, you need to keep track of the partial sum that needs to be updated every time a new number is given.The working memory is precisely what would make it possible to maintain (i.e. temporarily store) the partial sum and to update it (i.e. manipulate).In this thesis, we propose to explore the neuronal implementations of this working memory using a limited number of hypotheses.To do this, we place ourselves in the general context of recurrent neural networks and we propose to use in particular the reservoir computing paradigm.This type of very simple model nevertheless makes it possible to produce dynamics that learning can take advantage of to solve a given task.In this job, the task to be performed is a gated working memory task.The model receives as input a signal which controls the update of the memory.When the door is closed, the model should maintain its current memory state, while when open, it should update it based on an input.In our approach, this additional input is present at all times, even when there is no update to do.In other words, we require our model to be an open system, i.e. a system which is always disturbed by its inputs but which must nevertheless learn to keep a stable memory.In the first part of this work, we present the architecture of the model and its properties, then we show its robustness through a parameter sensitivity study.This shows that the model is extremely robust for a wide range of parameters.More or less, any random population of neurons can be used to perform gating.Furthermore, after learning, we highlight an interesting property of the model, namely that information can be maintained in a fully distributed manner, i.e. without being correlated to any of the neurons but only to the dynamics of the group.More precisely, working memory is not correlated with the sustained activity of neurons, which has nevertheless been observed for a long time in the literature and recently questioned experimentally.This model confirms these results at the theoretical level.In the second part of this work, we show how these models obtained by learning can be extended in order to manipulate the information which is in the latent space.We therefore propose to consider conceptors which can be conceptualized as a set of synaptic weights which constrain the dynamics of the reservoir and direct it towards particular subspaces; for example subspaces corresponding to the maintenance of a particular value.More generally, we show that these conceptors can not only maintain information, they can also maintain functions.In the case of mental arithmetic mentioned previously, these conceptors then make it possible to remember and apply the operation to be carried out on the various inputs given to the system.These conceptors therefore make it possible to instantiate a procedural working memory in addition to the declarative working memory.We conclude this work by putting this theoretical model into perspective with respect to biology and neurosciences
Etienne, Caroline. "Apprentissage profond appliqué à la reconnaissance des émotions dans la voix". Thesis, Université Paris-Saclay (ComUE), 2019. http://www.theses.fr/2019SACLS517.
Pełny tekst źródłaThis thesis deals with the application of artificial intelligence to the automatic classification of audio sequences according to the emotional state of the customer during a commercial phone call. The goal is to improve on existing data preprocessing and machine learning models, and to suggest a model that is as efficient as possible on the reference IEMOCAP audio dataset. We draw from previous work on deep neural networks for automatic speech recognition, and extend it to the speech emotion recognition task. We are therefore interested in End-to-End neural architectures to perform the classification task including an autonomous extraction of acoustic features from the audio signal. Traditionally, the audio signal is preprocessed using paralinguistic features, as part of an expert approach. We choose a naive approach for data preprocessing that does not rely on specialized paralinguistic knowledge, and compare it with the expert approach. In this approach, the raw audio signal is transformed into a time-frequency spectrogram by using a short-term Fourier transform. In order to apply a neural network to a prediction task, a number of aspects need to be considered. On the one hand, the best possible hyperparameters must be identified. On the other hand, biases present in the database should be minimized (non-discrimination), for example by adding data and taking into account the characteristics of the chosen dataset. We study these aspects in order to develop an End-to-End neural architecture that combines convolutional layers specialized in the modeling of visual information with recurrent layers specialized in the modeling of temporal information. We propose a deep supervised learning model, competitive with the current state-of-the-art when trained on the IEMOCAP dataset, justifying its use for the rest of the experiments. This classification model consists of a four-layer convolutional neural networks and a bidirectional long short-term memory recurrent neural network (BLSTM). Our model is evaluated on two English audio databases proposed by the scientific community: IEMOCAP and MSP-IMPROV. A first contribution is to show that, with a deep neural network, we obtain high performances on IEMOCAP, and that the results are promising on MSP-IMPROV. Another contribution of this thesis is a comparative study of the output values of the layers of the convolutional module and the recurrent module according to the data preprocessing method used: spectrograms (naive approach) or paralinguistic indices (expert approach). We analyze the data according to their emotion class using the Euclidean distance, a deterministic proximity measure. We try to understand the characteristics of the emotional information extracted autonomously by the network. The idea is to contribute to research focused on the understanding of deep neural networks used in speech emotion recognition and to bring more transparency and explainability to these systems, whose decision-making mechanism is still largely misunderstood
Jouffroy, Guillaume. "Contrôle oscillatoire par réseau de neurones récurrents". Paris 8, 2008. http://www.theses.fr/2008PA082918.
Pełny tekst źródłaIn the control field, most of the applications need a non-oscillatory continuous control. This work focuses instead on controllers with recurrent neural networks (RNN) which generate a periodic oscillatory control. The purpose of the present work is to study stochastic optimisation methods which can be used to discover the parameters of a network so that it generates a cyclic input. First we take a look at the knowledge about biological oscillators. Tthen we describe the mathematical tools to be able to guarantee the stability oscillators. The potential of RNN, especially applied to dynamical systems being still poorly used, we propose for each method, a general detailed matrix formalization and we precise the computational complexity of the methods. We validate each method using a simple example of oscillator, and we demonstrate analytically the stability of the resulting oscillator, but also how it is robust to parameters perturbations. We then compare these different methods with these criteria and the speed of convergence. We finish this thesis with an illustration, where we take all the steps of the construction of an oscillatory neural controller, to control the axis of direction of a particular vehicle. This will let us discuss how realistic is the use of recurrent neural networks in the field of control, and propose interesting questions
Jodouin, Jean-François. "Réseaux de neurones et traitement du langage naturel : étude des réseaux de neurones récurrents et de leurs représentations". Paris 11, 1993. http://www.theses.fr/1993PA112079.
Pełny tekst źródłaDaucé, Emmanuel. "Adaptation dynamique et apprentissage dans les réseaux de neurones récurrents aléatoires". Toulouse, ENSAE, 2000. https://tel.archives-ouvertes.fr/tel-01394004.
Pełny tekst źródłaStuner, Bruno. "Cohorte de réseaux de neurones récurrents pour la reconnaissance de l'écriture". Thesis, Normandie, 2018. http://www.theses.fr/2018NORMR024.
Pełny tekst źródłaState-of-the-art methods for handwriting recognition are based on LSTM recurrent neural networks (RNN) which achieve high performance recognition. In this thesis, we propose the lexicon verification and the cohort generation as two new building blocs to tackle the problem of handwriting recognition which are : i) the large vocabulary problem and the use of lexicon driven methods ii) the combination of multiple optical models iii) the need for large labeled dataset for training RNN. The lexicon verification is an alternative to the lexicon driven decoding process and can deal with lexicons of 3 millions words. The cohort generation is a method to get easily and quickly a large number of complementary recurrent neural networks extracted from a single training. From these two new techniques we build and propose a new cascade scheme for isolated word recognition, a new line level combination LV-ROVER and a new self-training strategy to train LSTM RNN for isolated handwritten words recognition. The proposed cascade combines thousands of LSTM RNN with lexicon verification and achieves state-of-the art word recognition performance on the Rimes and IAM datasets. The Lexicon Verified ROVER : LV-ROVER, has a reduce complexity compare to the original ROVER algorithm and combine hundreds of recognizers without language models while achieving state of the art for handwritten line text on the RIMES dataset. Our self-training strategy use both labeled and unlabeled data with the unlabeled data being self-labeled by its own lexicon verified predictions. The strategy enables self-training with a single BLSTM and show excellent results on the Rimes and Iam datasets
Abbasi, Mahdieh. "Toward robust deep neural networks". Doctoral thesis, Université Laval, 2020. http://hdl.handle.net/20.500.11794/67766.
Pełny tekst źródłaIn this thesis, our goal is to develop robust and reliable yet accurate learning models, particularly Convolutional Neural Networks (CNNs), in the presence of adversarial examples and Out-of-Distribution (OOD) samples. As the first contribution, we propose to predict adversarial instances with high uncertainty through encouraging diversity in an ensemble of CNNs. To this end, we devise an ensemble of diverse specialists along with a simple and computationally efficient voting mechanism to predict the adversarial examples with low confidence while keeping the predictive confidence of the clean samples high. In the presence of high entropy in our ensemble, we prove that the predictive confidence can be upper-bounded, leading to have a globally fixed threshold over the predictive confidence for identifying adversaries. We analytically justify the role of diversity in our ensemble on mitigating the risk of both black-box and white-box adversarial examples. Finally, we empirically assess the robustness of our ensemble to the black-box and the white-box attacks on several benchmark datasets.The second contribution aims to address the detection of OOD samples through an end-to-end model trained on an appropriate OOD set. To this end, we address the following central question: how to differentiate many available OOD sets w.r.t. a given in distribution task to select the most appropriate one, which in turn induces a model with a high detection rate of unseen OOD sets? To answer this question, we hypothesize that the “protection” level of in-distribution sub-manifolds by each OOD set can be a good possible property to differentiate OOD sets. To measure the protection level, we then design three novel, simple, and cost-effective metrics using a pre-trained vanilla CNN. In an extensive series of experiments on image and audio classification tasks, we empirically demonstrate the abilityof an Augmented-CNN (A-CNN) and an explicitly-calibrated CNN for detecting a significantly larger portion of unseen OOD samples, if they are trained on the most protective OOD set. Interestingly, we also observe that the A-CNN trained on the most protective OOD set (calledA-CNN) can also detect the black-box Fast Gradient Sign (FGS) adversarial examples. As the third contribution, we investigate more closely the capacity of the A-CNN on the detection of wider types of black-box adversaries. To increase the capability of A-CNN to detect a larger number of adversaries, we augment its OOD training set with some inter-class interpolated samples. Then, we demonstrate that the A-CNN trained on the most protective OOD set along with the interpolated samples has a consistent detection rate on all types of unseen adversarial examples. Where as training an A-CNN on Projected Gradient Descent (PGD) adversaries does not lead to a stable detection rate on all types of adversaries, particularly the unseen types. We also visually assess the feature space and the decision boundaries in the input space of a vanilla CNN and its augmented counterpart in the presence of adversaries and the clean ones. By a properly trained A-CNN, we aim to take a step toward a unified and reliable end-to-end learning model with small risk rates on both clean samples and the unusual ones, e.g. adversarial and OOD samples.The last contribution is to show a use-case of A-CNN for training a robust object detector on a partially-labeled dataset, particularly a merged dataset. Merging various datasets from similar contexts but with different sets of Object of Interest (OoI) is an inexpensive way to craft a large-scale dataset which covers a larger spectrum of OoIs. Moreover, merging datasets allows achieving a unified object detector, instead of having several separate ones, resultingin the reduction of computational and time costs. However, merging datasets, especially from a similar context, causes many missing-label instances. With the goal of training an integrated robust object detector on a partially-labeled but large-scale dataset, we propose a self-supervised training framework to overcome the issue of missing-label instances in the merged datasets. Our framework is evaluated on a merged dataset with a high missing-label rate. The empirical results confirm the viability of our generated pseudo-labels to enhance the performance of YOLO, as the current (to date) state-of-the-art object detector.
Galtier, Mathieu. "Une approche mathématique de l'apprentissage non-supervisé dans les réseaux de neurones récurrents". Phd thesis, École Nationale Supérieure des Mines de Paris, 2011. http://pastel.archives-ouvertes.fr/pastel-00667368.
Pełny tekst źródłaFourure, Damien. "Réseaux de neurones convolutifs pour la segmentation sémantique et l'apprentissage d'invariants de couleur". Thesis, Lyon, 2017. http://www.theses.fr/2017LYSES056/document.
Pełny tekst źródłaComputer vision is an interdisciplinary field that investigates how computers can gain a high level of understanding from digital images or videos. In artificial intelligence, and more precisely in machine learning, the field in which this thesis is positioned,computer vision involves extracting characteristics from images and then generalizing concepts related to these characteristics. This field of research has become very popular in recent years, particularly thanks to the results of the convolutional neural networks that form the basis of so-called deep learning methods. Today, neural networks make it possible, among other things, to recognize different objects present in an image, to generate very realistic images or even to beat the champions at the Go game. Their performance is not limited to the image domain, since they are also used in other fields such as natural language processing (e. g. machine translation) or sound recognition. In this thesis, we study convolutional neural networks in order to develop specialized architectures and loss functions for low-level tasks (color constancy) as well as high-level tasks (semantic segmentation). Color constancy, is the ability of the human visual system to perceive constant colours for a surface despite changes in the spectrum of illumination (lighting change). In computer vision, the main approach consists in estimating the color of the illuminant and then suppressing its impact on the perceived color of objects. We approach the task of color constancy with the use of neural networks by developing a new architecture composed of a subsampling operator inspired by traditional methods. Our experience shows that our method makes it possible to obtain competitive performances with the state of the art. Nevertheless, our architecture requires a large amount of training data. In order to partially correct this problem and improve the training of neural networks, we present several techniques for artificial data augmentation. We are also making two contributions on a high-level issue : semantic segmentation. This task, which consists of assigning a semantic class to each pixel of an image, is a challenge in computer vision because of its complexity. On the one hand, it requires many examples of training that are costly to obtain. On the other hand, it requires the adaptation of traditional convolutional neural networks in order to obtain a so-called dense prediction, i. e., a prediction for each pixel present in the input image. To solve the difficulty of acquiring training data, we propose an approach that uses several databases annotated with different labels at the same time. To do this, we define a selective loss function that has the advantage of allowing the training of a convolutional neural network from data from multiple databases. We also developed self-context approach that captures the correlations between labels in different databases. Finally, we present our third contribution : a new convolutional neural network architecture called GridNet specialized for semantic segmentation. Unlike traditional networks, implemented with a single path from the input (image) to the output (prediction), our architecture is implemented as a 2D grid allowing several interconnected streams to operate at different resolutions. In order to exploit all the paths of the grid, we propose a technique inspired by dropout. In addition, we empirically demonstrate that our architecture generalize many of well-known stateof- the-art networks. We conclude with an analysis of the empirical results obtained with our architecture which, although trained from scratch, reveals very good performances, exceeding popular approaches often pre-trained
Bouaziz, Mohamed. "Réseaux de neurones récurrents pour la classification de séquences dans des flux audiovisuels parallèles". Thesis, Avignon, 2017. http://www.theses.fr/2017AVIG0224/document.
Pełny tekst źródłaIn the same way as TV channels, data streams are represented as a sequence of successive events that can exhibit chronological relations (e.g. a series of programs, scenes, etc.). For a targeted channel, broadcast programming follows the rules defined by the channel itself, but can also be affected by the programming of competing ones. In such conditions, event sequences of parallel streams could provide additional knowledge about the events of a particular stream. In the sphere of machine learning, various methods that are suited for processing sequential data have been proposed. Long Short-Term Memory (LSTM) Recurrent Neural Networks have proven its worth in many applications dealing with this type of data. Nevertheless, these approaches are designed to handle only a single input sequence at a time. The main contribution of this thesis is about developing approaches that jointly process sequential data derived from multiple parallel streams. The application task of our work, carried out in collaboration with the computer science laboratory of Avignon (LIA) and the EDD company, seeks to predict the genre of a telecast. This prediction can be based on the histories of previous telecast genres in the same channel but also on those belonging to other parallel channels. We propose a telecast genre taxonomy adapted to such automatic processes as well as a dataset containing the parallel history sequences of 4 French TV channels. Two original methods are proposed in this work in order to take into account parallel stream sequences. The first one, namely the Parallel LSTM (PLSTM) architecture, is an extension of the LSTM model. PLSTM simultaneously processes each sequence in a separate recurrent layer and sums the outputs of each of these layers to produce the final output. The second approach, called MSE-SVM, takes advantage of both LSTM and Support Vector Machines (SVM) methods. Firstly, latent feature vectors are independently generated for each input stream, using the output event of the main one. These new representations are then merged and fed to an SVM algorithm. The PLSTM and MSE-SVM approaches proved their ability to integrate parallel sequences by outperforming, respectively, the LSTM and SVM models that only take into account the sequences of the main stream. The two proposed approaches take profit of the information contained in long sequences. However, they have difficulties to deal with short ones. Though MSE-SVM generally outperforms the PLSTM approach, the problem experienced with short sequences is more pronounced for MSE-SVM. Finally, we propose to extend this approach by feeding additional information related to each event in the input sequences (e.g. the weekday of a telecast). This extension, named AMSE-SVM, has a remarkably better behavior with short sequences without affecting the performance when processing long ones
Suzano, Massa Francisco Vitor. "Mise en relation d'images et de modèles 3D avec des réseaux de neurones convolutifs". Thesis, Paris Est, 2017. http://www.theses.fr/2017PESC1198/document.
Pełny tekst źródłaThe recent availability of large catalogs of 3D models enables new possibilities for a 3D reasoning on photographs. This thesis investigates the use of convolutional neural networks (CNNs) for relating 3D objects to 2D images.We first introduce two contributions that are used throughout this thesis: an automatic memory reduction library for deep CNNs, and a study of CNN features for cross-domain matching. In the first one, we develop a library built on top of Torch7 which automatically reduces up to 91% of the memory requirements for deploying a deep CNN. As a second point, we study the effectiveness of various CNN features extracted from a pre-trained network in the case of images from different modalities (real or synthetic images). We show that despite the large cross-domain difference between rendered views and photographs, it is possible to use some of these features for instance retrieval, with possible applications to image-based rendering.There has been a recent use of CNNs for the task of object viewpoint estimation, sometimes with very different design choices. We present these approaches in an unified framework and we analyse the key factors that affect performance. We propose a joint training method that combines both detection and viewpoint estimation, which performs better than considering the viewpoint estimation separately. We also study the impact of the formulation of viewpoint estimation either as a discrete or a continuous task, we quantify the benefits of deeper architectures and we demonstrate that using synthetic data is beneficial. With all these elements combined, we improve over previous state-of-the-art results on the Pascal3D+ dataset by a approximately 5% of mean average viewpoint precision.In the instance retrieval study, the image of the object is given and the goal is to identify among a number of 3D models which object it is. We extend this work to object detection, where instead we are given a 3D model (or a set of 3D models) and we are asked to locate and align the model in the image. We show that simply using CNN features are not enough for this task, and we propose to learn a transformation that brings the features from the real images close to the features from the rendered views. We evaluate our approach both qualitatively and quantitatively on two standard datasets: the IKEAobject dataset, and a subset of the Pascal VOC 2012 dataset of the chair category, and we show state-of-the-art results on both of them
Low, Kok Seng. "Approche de réseaux de neurones récurrents pour le problème de l'ordonnancement cyclique et sa variante". Artois, 2008. http://www.theses.fr/2008ARTO0408.
Pełny tekst źródłaScheduling deals with the allocation of required tasks to limited resources over time, to be processed. The scheduling problems arise among others, in areas of product manufacturing, computer processing and transportation. In this thesis we focus on the cyclic version of the scheduling problem. We review the properties of both the general scheduling and cyclic scheduling problems. As the cyclic scheduling problem is NP-Hard complexity, the time to solve the problem requires exponential time in the worst case scenario. This factor has motivated this research work in developing an efficient neural network approach to solve the cyclic scheduling problem. This thesis focuses specifically on the cyclic job shop and cyclic flexible manufacturing system problems hence models that will solve the minimum cycle time or work in progress of the problems, were developed. These models are fundamental to which the neural network approach can be applied. From the literature, the absence of neural network research into solving the scheduling problem is due to its characteristics such as complex architecture, defining initial conditions, difficulty in tuning its parameters (i. E. Learning rate, stoppage conditions, etc) and tendency for infeasible solutions. However, in this thesis, we develop and study three variations of the neural network approaches. These are the Recurrent Neural Network (RNN) approach, the Lagrangian Relaxation Recurrent Neural Network (LRRNN) approach and the Advanced Hopfield network approach. Several algorithms were combined with these neural networks to ensure that feasible solutions are generated and to reduce the search effort for the optimum solutions. A Competitive Dispatch Rule Phase (CDRP) was developed to generate initial feasible solutions before the three neural network approaches are initiated. This is important as the search space of the problem can be reduced through this approach. For the cyclic flexible manufacturing system problem, a Modified Competitive Dispatch Rule Phase (MCDRP) is developed in response to having the best possible cyclic schedule with minimum work in progress, for the neural network approaches to work from. As the solutions may be trapped in local minimum energy state, a schedule perturbation phase was developed to ''kick-start'' the search effort. Finally using the developed schedule Postprocessing phase that contains the Adhere Conjunctive and Adhere Disjunctive algorithms, the subsequent final solutions are always feasible schedules. We also extended the review into the cyclic job shop problem with linear precedence constraints. From cyclic scheduling literature, it is possible to transform the linear constraints into the equivalent uniform forms, hence the Delinearization algorithm was developed. We were able to demonstrate the suitability and applicability of the RNN, LRRNN and Advanced Hopfield network approaches through computational and comparative testing. The experimental results indicate that the three approaches are attractive alternatives to traditional heuristics in solving the cyclic scheduling problems, even though in some cases, it is computational expensive
Nono, Wouafo Hugues Gérald. "Architectures matérielles numériques intégrées et réseaux de neurones à codage parcimonieux". Thesis, Lorient, 2016. http://www.theses.fr/2016LORIS394/document.
Pełny tekst źródłaNowadays, artificial neural networks are widely used in many applications such as image and signal processing. Recently, a new model of neural network was proposed to design associative memories, the GBNN (Gripon-Berrou Neural Network). This model offers a storage capacity exceeding those of Hopfield networks when the information to be stored has a uniform distribution. Methods improving performance for non-uniform distributions and hardware architectures implementing the GBNN networks were proposed. However, on one hand, these solutions are very expensive in terms of hardware resources and on the other hand, the proposed architectures can only implement fixed size networks and are not scalable. The objectives of this thesis are: (1) to design GBNN inspired models outperforming the state of the art, (2) to propose architectures cheaper than existing solutions and (3) to design a generic architecture implementing the proposed models and able to handle various sizes of networks. The results of these works are exposed in several parts. Initially, the concept of clone based neural networks and its variants are presented. These networks offer better performance than the state of the art for the same memory cost when a non-uniform distribution of the information to be stored is considered. The hardware architecture optimizations are then introduced to significantly reduce the cost in terms of resources. Finally, a generic scalable architecture able to handle various sizes of networks is proposed
Chabot, Florian. "Analyse fine 2D/3D de véhicules par réseaux de neurones profonds". Thesis, Université Clermont Auvergne (2017-2020), 2017. http://www.theses.fr/2017CLFAC018/document.
Pełny tekst źródłaIn this thesis, we are interested in fine-grained analysis of vehicle from an image. We define fine-grained analysis as the following concepts : vehicle detection in the image, vehicle viewpoint (or orientation) estimation, vehicle visibility characterization, vehicle 3D localization and make and model recognition. The design of reliable solutions for fine-grained analysis of vehicle open the door to multiple applications in particular for intelligent transport systems as well as video surveillance systems. In this work, we propose several contributions allowing to address partially or wholly this issue. Proposed approaches are based on joint deep learning technologies and 3D models. In a first section, we deal with make and model classification keeping in mind the difficulty to create training data. In a second section, we investigate a novel method for both vehicle detection and fine-grained viewpoint estimation based on local apparence features and geometric spatial coherence. It uses models learned only on synthetic data. Finally, in a third section, a complete system for fine-grained analysis is proposed. It is based on the multi-task concept. Throughout this report, we provide quantitative and qualitative results. On several aspects related to vehicle fine-grained analysis, this work allowed to outperform state of the art methods
Assaad, Mohammad. "Un nouvel algorithme de boosting pour les réseaux de neurones récurrents : application au traitement des données sequentielles". Tours, 2006. http://www.theses.fr/2006TOUR4024.
Pełny tekst źródłaThe work of this thesis deals with the proposal of a new boosting algorithm dedicated to the problem of learning time-dependencies for the time series prediction, using recurrent neural networks as regressors. This algorithm is based on the boosting algorith and allows concentrating the training on difficult examples. A new parameter is introduced to regulate the influence of boosting. To evaluate our algorithm, systematic experiments were carried out on two types of problems of time series prediction : single-step ahead predicton and multi-step ahead prediction. The results obtained from several series of reference are close to the best results reported in the literature
Farabet, Clément. "Analyse sémantique des images en temps-réel avec des réseaux convolutifs". Phd thesis, Université Paris-Est, 2013. http://tel.archives-ouvertes.fr/tel-00965622.
Pełny tekst źródłaGelly, Grégory. "Réseaux de neurones récurrents pour le traitement automatique de la parole". Thesis, Université Paris-Saclay (ComUE), 2017. http://www.theses.fr/2017SACLS295/document.
Pełny tekst źródłaAutomatic speech processing is an active field of research since the 1950s. Within this field the main area of research is automatic speech recognition but simpler tasks such as speech activity detection, language identification or speaker identification are also of great interest to the community. The most recent breakthrough in speech processing appeared around 2010 when speech recognition systems using deep neural networks drastically improved the state-of-the-art. Inspired by this gains and the work of Alex Graves on recurrent neural networks (RNN), we decided to explore the possibilities brought by these models on realistic data for two different tasks: speech activity detection and spoken language identification. In this work, we closely look at a specific model for the RNNs: the Long Short Term Memory (LSTM) which mitigates a lot of the difficulties that can arise when training an RNN. We augment this model and introduce optimization methods that lead to significant performance gains for speech activity detection and language identification. More specifically, we introduce a WER-like loss function to train a speech activity detection system so as to minimize the word error rate of a downstream speech recognition system. We also introduce two different methods to successfully train a multiclass classifier based on neural networks for tasks such as LID. The first one is based on a divide-and-conquer approach and the second one is based on an angular proximity loss function. Both yield performance gains but also speed up the training process
Thiaw, Lamine. "Identification de systèmes dynamiques non linéaires par réseaux de neurones et multimodèles". Phd thesis, Université Paris XII Val de Marne, 2008. http://tel.archives-ouvertes.fr/tel-00399469.
Pełny tekst źródłaMorillot, Olivier. "Reconnaissance de textes manuscrits par modèles de Markov cachés et réseaux de neurones récurrents : application à l'écriture latine et arabe". Electronic Thesis or Diss., Paris, ENST, 2014. http://www.theses.fr/2014ENST0002.
Pełny tekst źródłaHandwriting recognition is an essential component of document analysis. One of the popular trends is to go from isolated word to word sequence recognition. Our work aims to propose a text-line recognition system without explicit word segmentation. In order to build an efficient model, we intervene at different levels of the recognition system. First of all, we introduce two new preprocessing techniques : a cleaning and a local baseline correction for text-lines. Then, a language model is built and optimized for handwritten mails. Afterwards, we propose two state-of-the-art recognition systems based on contextual HMMs (Hidden Markov Models) and recurrent neural networks BLSTM (Bi-directional Long Short-Term Memory). We optimize our systems in order to give a comparison of those two approaches. Our systems are evaluated on arabic and latin cursive handwritings and have been submitted to two international handwriting recognition competitions. At last, we introduce a strategy for some out-of-vocabulary character strings recognition, as a prospect of future work
Beltzung, Benjamin. "Utilisation de réseaux de neurones convolutifs pour mieux comprendre l’évolution et le développement du comportement de dessin chez les Hominidés". Electronic Thesis or Diss., Strasbourg, 2023. http://www.theses.fr/2023STRAJ114.
Pełny tekst źródłaThe study of drawing behavior can be highly informative, both cognitively and psychologically, in humans and other primates. However, this wealth of information can also be a challenge to analysis and interpretation, particularly in the absence of explanation or verbalization by the author of the drawing. Indeed, an adult's interpretation of a drawing may not be in line with the artist's original intention. During my thesis, I showed that, although generally regarded as black boxes, convolutional neural networks (CNNs) can provide a better understanding of the drawing behavior. Firstly, by using a CNN to classify drawings of a female orangutan according to their season of production, and highlighting variation in style and content. In addition, an ontogenetic approach was considered to quantify the similarity between productions from different age groups. In the future, more interpretable models and the application of new interpretability methods could be applied to better decipher drawing behavior
Chraibi, Kaadoud Ikram. "apprentissage de séquences et extraction de règles de réseaux récurrents : application au traçage de schémas techniques". Thesis, Bordeaux, 2018. http://www.theses.fr/2018BORD0032/document.
Pełny tekst źródłaThere are two important aspects of the knowledge that an individual acquires through experience. One corresponds to the semantic memory (explicit knowledge, such as the learning of concepts and categories describing the objects of the world) and the other, the procedural or syntactic memory (knowledge relating to the learning of rules or syntax). This "syntactic memory" is built from experience and particularly from the observation of sequences of objects whose organization obeys syntactic rules.It must have the capability to aid recognizing as well as generating valid sequences in the future, i.e., sequences respecting the learnt rules. This production of valid sequences can be done either in an explicit way, that is, by evoking the underlying rules, or implicitly, when the learning phase has made it possible to capture the principle of organization of the sequences without explicit recourse to the rules. Although the latter is faster, more robust and less expensive in terms of cognitive load as compared to explicit reasoning, the implicit process has the disadvantage of not giving access to the rules and thus becoming less flexible and less explicable. These mnemonic mechanisms can also be applied to business expertise. The capitalization of information and knowledge in general, for any company is a major issue and concerns both the explicit and implicit knowledge. At first, the expert makes a choice to explicitly follow the rules of the trade. But then, by dint of repetition, the choice is made automatically, without explicit evocation of the underlying rules. This change in encoding rules in an individual in general and particularly in a business expert can be problematic when it is necessary to explain or transmit his or her knowledge. Indeed, if the business concepts can be formalized, it is usually in any other way for the expertise which is more difficult to extract and transmit.In our work, we endeavor to observe sequences of electrical components and in particular the problem of extracting rules hidden in these sequences, which are an important aspect of the extraction of business expertise from technical drawings. We place ourselves in the connectionist domain, and we have particularly considered neuronal models capable of processing sequences. We implemented two recurrent neural networks: the Elman model and a model with LSTM (Long Short Term Memory) units. We have evaluated these two models on different artificial grammars (Reber's grammar and its variations) in terms of learning, their generalization abilities and their management of sequential dependencies. Finally, we have also shown that it is possible to extract the encoded rules (from the sequences) in the recurrent network with LSTM units, in the form of an automaton. The electrical domain is particularly relevant for this problem. It is more constrained with a limited combinatorics than the planning of tasks in general cases like navigation for example, which could constitute a perspective of this work
Çinar, Yagmur Gizem. "Prédiction de séquences basée sur des réseaux de neurones récurrents dans le contexte des séries temporelles et des sessions de recherche d'information". Thesis, Université Grenoble Alpes (ComUE), 2019. http://www.theses.fr/2019GREAM079.
Pełny tekst źródłaThis thesis investigates challenges of sequence prediction in different scenarios such as sequence prediction using recurrent neural networks (RNNs) in the context of time series and information retrieval (IR) search sessions. Predicting the unknown values that follow some previously observed values is basically called sequence prediction.It is widely applicable to many domains where a sequential behavior is observed in the data. In this study, we focus on two different types of sequence prediction tasks: time series forecasting and next query prediction in an information retrieval search session.Time series often display pseudo-periods, i.e. time intervals with strong correlation between values of time series. Seasonal changes in weather time series or electricity usage at day and night time are some examples of pseudo-periods. In a forecasting scenario, pseudo-periods correspond to the difference between the positions of the output being predicted and specific inputs.In order to capture periods in RNNs, one needs a memory of the input sequence. Sequence-to-sequence RNNs (with attention mechanism) reuse specific (representations of) input values to predict output values. Sequence-to-sequence RNNs with an attention mechanism seem to be adequate for capturing periods. In this manner, we first explore the capability of an attention mechanism in that context. However, according to our initial analysis, a standard attention mechanism did not perform well to capture the periods. Therefore, we propose a period-aware content-based attention RNN model. This model is an extension of state-of-the-art sequence-to-sequence RNNs with attention mechanism and it is aimed to capture the periods in time series with or without missing values.Our experimental results with period-aware content-based attention RNNs show significant improvement on univariate and multivariate time series forecasting performance on several publicly available data sets.Another challenge in sequence prediction is the next query prediction. The next query prediction helps users to disambiguate their search query, to explore different aspects of the information they need or to form a precise and succint query that leads to higher retrieval performance. A search session is dynamic, and the information need of a user might change over a search session as a result of the search interactions. Furthermore, interactions of a user with a search engine influence the user's query reformulations. Considering this influence on the query formulations, we first analyze where the next query words come from? Using the analysis of the sources of query words, we propose two next query prediction approaches: a set view and a sequence view.The set view adapts a bag-of-words approach using a novel feature set defined based on the sources of next query words analysis. Here, the next query is predicted using learning to rank. The sequence view extends a hierarchical RNN model by considering the sources of next query words in the prediction. The sources of next query words are incorporated by using an attention mechanism on the interaction words. We have observed using sequence approach, a natural formulation of the problem, and exploiting all sources of evidence lead to better next query prediction
Aussem, Alexandre. "Théorie et applications des réseaux de neurones récurrents et dynamiques à la prédiction, à la modélisation et au contrôle adaptif des processus dynamiques". Paris 5, 1995. http://www.theses.fr/1995PA05S002.
Pełny tekst źródłaZEMOURI, RYAD. "Contribution à la surveillance des systèmes de production à l'aide des réseaux de neurones dynamiques : Application à la e-maintenance". Phd thesis, Université de Franche-Comté, 2003. http://tel.archives-ouvertes.fr/tel-00006003.
Pełny tekst źródłaBourdoukan, Ralph. "Le rôle de la balance entre excitation et inhibition dans l'apprentissage dans les réseaux de neurones à spikes". Thesis, Paris 6, 2016. http://www.theses.fr/2016PA066712/document.
Pełny tekst źródłaWhen performing a task, neural circuits must represent and manipulate continuous stimuli using discrete action potentials. It is commonly assumed that neurons represent continuous quantities with their firing rate and this independently from one another. However, such independent coding is very inefficient because it requires the generation of a large number of action potentials in order to achieve a certain level of accuracy. We show that neurons in a spiking recurrent network can learn - using a local plasticity rule - to coordinate their action potentials in order to represent information with high accuracy while discharging minimally. The learning rule that acts on recurrent connections leads to such an efficient coding by imposing a precise balance between excitation and inhibition at the level of each neuron. This balance is a frequently observed phenomenon in the brain and is central in our work. We also derive two biologically plausible learning rules that respectively allows the network to adapt to the statistics of its inputs and to perform complex and dynamic transformations on them. Finally, in these networks, the stochasticity of the spike timing is not a signature of noise but rather of precision and efficiency. In fact, the random nature of the spike times results from the degeneracy of the representation. This constitutes a new and a radically different interpretation of the irregularity found in spike trains
Boitard, Constance. "Identification des réseaux neurobiologiques gouvernant les apprentissages ambigus chez l'abeille Apis mellifera". Thesis, Toulouse 3, 2015. http://www.theses.fr/2015TOU30125/document.
Pełny tekst źródłaAssociative learning spans different levels of complexity, from simple tasks involving simple causal relationships between events, to ambiguous tasks, in which animals have to solve complex discriminations based on non-linear associative links. We focused on two protocols presenting a temporal or configural ambiguity at the level of stimulus contingencies in honey bees (\textit{Apis mellifera}). We performed selective blockades of GABAergic signalisation from recurrent feedback neurons in the mushroom bodies (MBs), higher-order insect brain structures associated with memory storage and retrieval, and found that this blockade within the MB calyces impaired both ambiguous learning tasks, although if did not affect simple conditioning counterparts. We suggest that the A3v cluster of the GABA feedback neurons innervating the MBs calyces are thus dispensable for simple learning, but are required for counteracting stimulus ambiguity in complex discriminations in honey bees
Rodriguez, Guillaume. "Modélisation des bases neuronales de la mémoire de travail paramétrique dans le cortex préfrontal". Thesis, Paris 6, 2016. http://www.theses.fr/2016PA066340/document.
Pełny tekst źródłaParametric working memory – the fundamental ability to maintain and manipulate quantitative information transiently – is critical to many core brain functions (perception, action, decision, behavioral control, cognition). Across neurons of the prefrontal cortex, parametric working memory is expressed through persistent graded activities (multistability) encoding the amplitude of past quantitative information (e.g. a psychophysical quantity, a number of items). The causal origin of this multistability remains unclear. Using biophysical and analytical models, I first studied the mnemonic properties of individual neurons endowed with supraliminar conductances. I then evaluated the possible role of these properties in maintaining persistent graded activities in prefrontal recurrent networks. These realistic models suggest 1) the existence of a flexible form of cellular bistability, conditioned to the historical regulation of the intrinsic properties and the nature of the stimulation and 2) that this cellular bistability could participate, in interaction with synaptic reverberation, to the emergence of persistent graded collective dynamics in prefrontal networks, the neural correlate of parametric working memory
Biasutto-Lervat, Théo. "Modélisation de la coarticulation multimodale : vers l'animation d'une tête parlante intelligible". Electronic Thesis or Diss., Université de Lorraine, 2021. http://www.theses.fr/2021LORR0019.
Pełny tekst źródłaThis thesis deals with neural network based coarticulation modeling, and aims to synchronize facial animation of a 3D talking head with speech. Predicting articulatory movements is not a trivial task, as it is well known that production of a phoneme is greatly affected by its phonetic context, a phoneme called coarticulation. We propose in this work a coarticulation model, i.e. a model able to predict spatial trajectories of articulators from speech. We rely on a sequential model, the recurrent neural networks, and more specifically the Gated Recurrent Units, which are able to consider the articulation dynamic as a central component of its modeling. Unfortunately, the typical amount of data in articulatory and audiovisual databases seems to be quite low for a deep learning approach. To overcome this difficulty, we propose to integrate articulatory knowledge into the networks during its initialization. The RNNs robustness allow uw to apply our coarticulation model to predict both face and tongue movements, in french and german for the face, and in english and german for the tongue. Evaluation has been conducted through objective measures of the trajectories, and through experiments to ensure a complete reach of critical articulatory targets. We also conducted a subjective evaluation to attest the perceptual quality of the predicted articulation once applied to our facial animation system. Finally, we analyzed the model after training to explore phonetic knowledges learned
Helson, Pascal. "Étude de la plasticité pour des neurones à décharge en interaction". Thesis, Université Côte d'Azur, 2021. http://www.theses.fr/2021COAZ4013.
Pełny tekst źródłaIn this thesis, we study a phenomenon that may be responsible for our memory capacity: the synaptic plasticity. It modifies the links between neurons over time. This phenomenon is stochastic: it is the result of a series of diverse and numerous chemical processes. The aim of the thesis is to propose a model of plasticity for interacting spiking neurons. The main difficulty is to find a model that satisfies the following conditions: it must be both consistent with the biological results of the field and simple enough to be studied mathematically and simulated with a large number of neurons.In a first step, from a rather simple model of plasticity, we study the learning of external signals by a neural network as well as the forgetting time of this signal when the network is subjected to other signals (noise). The mathematical analysis allows us to control the probability to misevaluate the signal. From this, we deduce explicit bounds on the time during which a given signal is kept in memory.Next, we propose a model based on stochastic rules of plasticity as a function of the occurrence time of the neural electrical discharges (Spike Timing Dependent Plasticity, STDP). This model is described by a Piecewise Deterministic Markov Process (PDMP). The long time behaviour of such a neural network is studied using a slow-fast analysis. In particular, sufficient conditions are established under which the process associated with synaptic weights is ergodic. Finally, we make the link between two levels of modelling: the microscopic and the macroscopic approaches. Starting from the dynamics presented at a microscopic level (neuron model and its interaction with other neurons), we derive an asymptotic dynamics which represents the evolution of a typical neuron and its incoming synaptic weights: this is the mean field analysis of the model. We thus condense the information on the dynamics of the weights and that of the neurons into a single equation, that of a typical neuron
Moinnereau, Marc-Antoine. "Encodage d'un signal audio dans un électroencéphalogramme". Mémoire, Université de Sherbrooke, 2017. http://hdl.handle.net/11143/10554.
Pełny tekst źródłaCabana, Tanguy. "Large deviations for the dynamics of heterogeneous neural networks". Thesis, Paris 6, 2016. http://www.theses.fr/2016PA066551/document.
Pełny tekst źródłaThis thesis addresses the rigorous derivation of mean-field results for the continuous time dynamics of heterogeneous large neural networks. In our models, we consider firing-rate neurons subject to additive noise. The network is fully connected, with highly random connectivity weights. Their variance scales as the inverse of the network size, and thus conserves a non-trivial role in the thermodynamic limit. Moreover, another heterogeneity is considered at the level of each neuron. It is interpreted as a spatial location. For biological relevance, a model considered includes delays, mean and variance of connections depending on the distance between cells. A second model considers interactions depending on the states of both neurons at play. This last case notably applies to Kuramoto's model of coupled oscillators. When the weights are independent Gaussian random variables, we show that the empirical measure of the neurons' states satisfies a large deviations principle, with a good rate function achieving its minimum at a unique probability measure, implying averaged convergence of the empirical measure and propagation of chaos. In certain cases, we also obtained quenched results. The limit is characterized through a complex non Markovian implicit equation in which the network interaction term is replaced by a non-local Gaussian process whose statistics depend on the solution over the whole neural field. We further demonstrate the universality of this limit, in the sense that neuronal networks with non-Gaussian interconnections but sub-Gaussian tails converge towards it. Moreover, we present a few numerical applications, and discuss possible perspectives
Buniet, Laurent. "Traitement automatique de la parole en milieu bruité : étude de modèles connexionnistes statiques et dynamiques". Phd thesis, Université Henri Poincaré - Nancy I, 1997. http://tel.archives-ouvertes.fr/tel-00629285.
Pełny tekst źródłaAbdelouahab, Kamel. "Reconfigurable hardware acceleration of CNNs on FPGA-based smart cameras". Thesis, Université Clermont Auvergne (2017-2020), 2018. http://www.theses.fr/2018CLFAC042/document.
Pełny tekst źródłaDeep Convolutional Neural Networks (CNNs) have become a de-facto standard in computer vision. This success came at the price of a high computational cost, making the implementation of CNNs, under real-time constraints, a challenging task.To address this challenge, the literature exploits the large amount of parallelism exhibited by these algorithms, motivating the use of dedicated hardware platforms. In power-constrained environments, such as smart camera nodes, FPGA-based processing cores are known to be adequate solutions in accelerating computer vision applications. This is especially true for CNN workloads, which have a streaming nature that suits well to reconfigurable hardware architectures.In this context, the following thesis addresses the problems of CNN mapping on FPGAs. In Particular, it aims at improving the efficiency of CNN implementations through two main optimization strategies; The first one focuses on the CNN model and parameters while the second one considers the hardware architecture and the fine-grain building blocks
Adam, Chloé. "Pattern Recognition in the Usage Sequences of Medical Apps". Thesis, Université Paris-Saclay (ComUE), 2019. http://www.theses.fr/2019SACLC027/document.
Pełny tekst źródłaRadiologists use medical imaging solutions on a daily basis for diagnosis. Improving user experience is a major line of the continuous effort to enhance the global quality and usability of software products. Monitoring applications enable to record the evolution of various software and system parameters during their use and in particular the successive actions performed by the users in the software interface. These interactions may be represented as sequences of actions. Based on this data, this work deals with two industrial topics: software crashes and software usability. Both topics imply on one hand understanding the patterns of use, and on the other developing prediction tools either to anticipate crashes or to dynamically adapt software interface according to users' needs. First, we aim at identifying crash root causes. It is essential in order to fix the original defects. For this purpose, we propose to use a binomial test to determine which type of patterns is the most appropriate to represent crash signatures. The improvement of software usability through customization and adaptation of systems to each user's specific needs requires a very good knowledge of how users use the software. In order to highlight the trends of use, we propose to group similar sessions into clusters. We compare 3 session representations as inputs of different clustering algorithms. The second contribution of our thesis concerns the dynamical monitoring of software use. We propose two methods -- based on different representations of input actions -- to address two distinct industrial issues: next action prediction and software crash risk detection. Both methodologies take advantage of the recurrent structure of LSTM neural networks to capture dependencies among our sequential data as well as their capacity to potentially handle different types of input representations for the same data
Dridi, Aicha. "A novel efficient time series deep learning approach using classification, prediction and reinforcement : energy and telecom use case". Electronic Thesis or Diss., Institut polytechnique de Paris, 2022. http://www.theses.fr/2022IPPAS010.
Pełny tekst źródłaThe massive growth of sensors (temperature, humidity, accelerometer, position sensor) and mobile devices (smartphones, tablets, smartwatches) increases the amount of data generated explosively. This immense amount of data can be collected and managed. The work carried out during this thesis aims first to propose an approach that deals with a specific type of data, which are time series. First, we used classification methods based on convolutional neural networks and multilayer perceptrons to extract the relevant information. We then used recurrent neural networks to make the predictions. We treated several time series data: energy, cellular, and GPS taxi track data. We also investigated several other methods like as semantic compression and transfer learning. The two described methods above allow us for the first to transmit only the weight of the neural networks, or if an anomaly is detected, send the anomalous data. Transfer learning allows us to make good predictions even if the data is missing or noisy. These methods allowed us to set up dynamic anomaly detection mechanisms. The objective of the last part of the thesis is to develop and implement a resource management solution having as input the result of the previous phases. We used several methods to implement this resource management solution, such as reinforcement learning, exact resolution, or recurrent neural networks. The first application is the implementation of an energy management system. The second application is the management of the deployment of drones to assist cellular networks when an anomaly occurs
Chen, Dexiong. "Modélisation de données structurées avec des machines profondes à noyaux et des applications en biologie computationnelle". Thesis, Université Grenoble Alpes, 2020. http://www.theses.fr/2020GRALM070.
Pełny tekst źródłaDeveloping efficient algorithms to learn appropriate representations of structured data, including sequences or graphs, is a major and central challenge in machine learning. To this end, deep learning has become popular in structured data modeling. Deep neural networks have drawn particular attention in various scientific fields such as computer vision, natural language understanding or biology. For instance, they provide computational tools for biologists to possibly understand and uncover biological properties or relationships among macromolecules within living organisms. However, most of the success of deep learning methods in these fields essentially relies on the guidance of empirical insights as well as huge amounts of annotated data. Exploiting more data-efficient models is necessary as labeled data is often scarce.Another line of research is kernel methods, which provide a systematic and principled approach for learning non-linear models from data of arbitrary structure. In addition to their simplicity, they exhibit a natural way to control regularization and thus to avoid overfitting.However, the data representations provided by traditional kernel methods are only defined by simply designed hand-crafted features, which makes them perform worse than neural networks when enough labeled data are available. More complex kernels inspired by prior knowledge used in neural networks have thus been developed to build richer representations and thus bridge this gap. Yet, they are less scalable. By contrast, neural networks are able to learn a compact representation for a specific learning task, which allows them to retain the expressivity of the representation while scaling to large sample size.Incorporating complementary views of kernel methods and deep neural networks to build new frameworks is therefore useful to benefit from both worlds.In this thesis, we build a general kernel-based framework for modeling structured data by leveraging prior knowledge from classical kernel methods and deep networks. Our framework provides efficient algorithmic tools for learning representations without annotations as well as for learning more compact representations in a task-driven way. Our framework can be used to efficiently model sequences and graphs with simple interpretation of predictions. It also offers new insights about designing more expressive kernels and neural networks for sequences and graphs
Caye, Daudt Rodrigo. "Convolutional neural networks for change analysis in earth observation images with noisy labels and domain shifts". Electronic Thesis or Diss., Institut polytechnique de Paris, 2020. http://www.theses.fr/2020IPPAT033.
Pełny tekst źródłaThe analysis of satellite and aerial Earth observation images allows us to obtain precise information over large areas. A multitemporal analysis of such images is necessary to understand the evolution of such areas. In this thesis, convolutional neural networks are used to detect and understand changes using remote sensing images from various sources in supervised and weakly supervised settings. Siamese architectures are used to compare coregistered image pairs and to identify changed pixels. The proposed method is then extended into a multitask network architecture that is used to detect changes and perform land cover mapping simultaneously, which permits a semantic understanding of the detected changes. Then, classification filtering and a novel guided anisotropic diffusion algorithm are used to reduce the effect of biased label noise, which is a concern for automatically generated large-scale datasets. Weakly supervised learning is also achieved to perform pixel-level change detection using only image-level supervision through the usage of class activation maps and a novel spatial attention layer. Finally, a domain adaptation method based on adversarial training is proposed, which succeeds in projecting images from different domains into a common latent space where a given task can be performed. This method is tested not only for domain adaptation for change detection, but also for image classification and semantic segmentation, which proves its versatility
Pham, Huy-Hieu. "Architectures d'apprentissage profond pour la reconnaissance d'actions humaines dans des séquences vidéo RGB-D monoculaires : application à la surveillance dans les transports publics". Thesis, Toulouse 3, 2019. http://www.theses.fr/2019TOU30145.
Pełny tekst źródłaThis thesis is dealing with automatic recognition of human actions from monocular RGB-D video sequences. Our main goal is to recognize which human actions occur in unknown videos. This problem is a challenging task due to a number of obstacles caused by the variability of the acquisition conditions, including the lighting, the position, the orientation and the field of view of the camera, as well as the variability of actions which can be performed differently, notably in terms of speed. To tackle these problems, we first review and evaluate the most prominent state-of-the-art techniques to identify the current state of human action recognition in videos. We then propose a new approach for skeleton-based action recognition using Deep Neural Networks (DNNs). Two key questions have been addressed. First, how to efficiently represent the spatio-temporal patterns of skeletal data for fully exploiting the capacity in learning high-level representations of Deep Convolutional Neural Networks (D-CNNs). Second, how to design a powerful D-CNN architecture that is able to learn discriminative features from the proposed representation for classification task. As a result, we introduce two new 3D motion representations called SPMF (Skeleton Posture-Motion Feature) and Enhanced-SPMF that encode skeleton poses and their motions into color images. For learning and classification tasks, we design and train different D-CNN architectures based on the Residual Network (ResNet), Inception-ResNet-v2, Densely Connected Convolutional Network (DenseNet) and Efficient Neural Architecture Search (ENAS) to extract robust features from color-coded images and classify them. Experimental results on various public and challenging human action recognition datasets (MSR Action3D, Kinect Activity Recognition Dataset, SBU Kinect Interaction, and NTU-RGB+D) show that the proposed approach outperforms current state-of-the-art. We also conducted research on the problem of 3D human pose estimation from monocular RGB video sequences and exploited the estimated 3D poses for recognition task. Specifically, a deep learning-based model called OpenPose is deployed to detect 2D human poses. A DNN is then proposed and trained for learning a 2D-to-3D mapping in order to map the detected 2D keypoints into 3D poses. Our experiments on the Human3.6M dataset verified the effectiveness of the proposed method. These obtained results allow opening a new research direction for human action recognition from 3D skeletal data, when the depth cameras are failing. In addition, we collect and introduce in this thesis, CEMEST database, a new RGB-D dataset depicting passengers' behaviors in public transport. It consists of 203 untrimmed real-world surveillance videos of realistic "normal" and "abnormal" events. We achieve promising results on CEMEST with the support of data augmentation and transfer learning techniques. This enables the construction of real-world applications based on deep learning for enhancing public transportation management services
Al, Chami Zahi. "Estimation de la qualité des données multimedia en temps réel". Thesis, Pau, 2021. http://www.theses.fr/2021PAUU3066.
Pełny tekst źródłaOver the past decade, data providers have been generating and streaming a large amount of data, including images, videos, audio, etc. In this thesis, we will be focusing on processing images since they are the most commonly shared between the users on the global inter-network. In particular, treating images containing faces has received great attention due to its numerous applications, such as entertainment and social media apps. However, several challenges could arise during the processing and transmission phase: firstly, the enormous number of images shared and produced at a rapid pace requires a significant amount of time to be processed and delivered; secondly, images are subject to a wide range of distortions during the processing, transmission, or combination of many factors that could damage the images’content. Two main contributions are developed. First, we introduce a Full-Reference Image Quality Assessment Framework in Real-Time, capable of:1) preserving the images’content by ensuring that some useful visual information can still be extracted from the output, and 2) providing a way to process the images in real-time in order to cope with the huge amount of images that are being received at a rapid pace. The framework described here is limited to processing those images that have access to their reference version (a.k.a Full-Reference). Secondly, we present a No-Reference Image Quality Assessment Framework in Real-Time. It has the following abilities: a) assessing the distorted image without having its distortion-free image, b) preserving the most useful visual information in the images before publishing, and c) processing the images in real-time, even though the No-Reference image quality assessment models are considered very complex. Our framework offers several advantages over the existing approaches, in particular: i. it locates the distortion in an image in order to directly assess the distorted parts instead of processing the whole image, ii. it has an acceptable trade-off between quality prediction accuracy and execution latency, andiii. it could be used in several applications, especially these that work in real-time. The architecture of each framework is presented in the chapters while detailing the modules and components of the framework. Then, a number of simulations are made to show the effectiveness of our approaches to solve our challenges in relation to the existing approaches
Haykal, Vanessa. "Modélisation des séries temporelles par apprentissage profond". Thesis, Tours, 2019. http://www.theses.fr/2019TOUR4019.
Pełny tekst źródłaTime 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)
Firmo, Drumond Thalita. "Apports croisées de l'apprentissage hiérarchique et la modélisation du système visuel : catégorisation d'images sur des petits corpus de données". Thesis, Bordeaux, 2020. https://tel.archives-ouvertes.fr/tel-03129189.
Pełny tekst źródłaDeep convolutional neural networks (DCNN) have recently protagonized a revolution in large-scale object recognition. They have changed the usual computer vision practices of hand-engineered features, with their ability to hierarchically learn representative features from data with a pertinent classifier. Together with hardware advances, they have made it possible to effectively exploit the ever-growing amounts of image data gathered online. However, in specific domains like healthcare and industrial applications, data is much less abundant, and expert labeling costs higher than those of general purpose image datasets. This scarcity scenario leads to this thesis' core question: can these limited-data domains profit from the advantages of DCNNs for image classification? This question has been addressed throughout this work, based on an extensive study of literature, divided in two main parts, followed by proposal of original models and mechanisms.The first part reviews object recognition from an interdisciplinary double-viewpoint. First, it resorts to understanding the function of vision from a biological stance, comparing and contrasting to DCNN models in terms of structure, function and capabilities. Second, a state-of-the-art review is established aiming to identify the main architectural categories and innovations in modern day DCNNs. This interdisciplinary basis fosters the identification of potential mechanisms - inspired both from biological and artificial structures — that could improve image recognition under difficult situations. Recurrent processing is a clear example: while not completely absent from the "deep vision" literature, it has mostly been applied to videos — due to their inherently sequential nature. From biology however it is clear such processing plays a role in refining our perception of a still scene. This theme is further explored through a dedicated literature review focused on recurrent convolutional architectures used in image classification.The second part carries on in the spirit of improving DCNNs, this time focusing more specifically on our central question: deep learning over small datasets. First, the work proposes a more detailed and precise discussion of the small sample problem and its relation to learning hierarchical features with deep models. This discussion is followed up by a structured view of the field, organizing and discussing the different possible paths towards adapting deep models to limited data settings. Rather than a raw listing, this review work aims to make sense out of the myriad of approaches in the field, grouping methods with similar intent or mechanism of action, in order to guide the development of custom solutions for small-data applications. Second, this study is complemented by an experimental analysis, exploring small data learning with the proposition of original models and mechanisms (previously published as a journal paper).In conclusion, it is possible to apply deep learning to small datasets and obtain good results, if done in a thoughtful fashion. On the data path, one shall try gather more information from additional related data sources if available. On the complexity path, architecture and training methods can be calibrated in order to profit the most from any available domain-specific side-information. Proposals concerning both of these paths get discussed in detail throughout this document. Overall, while there are multiple ways of reducing the complexity of deep learning with small data samples, there is no universal solution. Each method has its own drawbacks and practical difficulties and needs to be tailored specifically to the target perceptual task at hand
Swaileh, Wassim. "Des modèles de langage pour la reconnaissance de l'écriture manuscrite". Thesis, Normandie, 2017. http://www.theses.fr/2017NORMR024/document.
Pełny tekst źródłaThis thesis is about the design of a complete processing chain dedicated to unconstrained handwriting recognition. Three main difficulties are adressed: pre-processing, optical modeling and language modeling. The pre-processing stage is related to extracting properly the text lines to be recognized from the document image. An iterative text line segmentation method using oriented steerable filters was developed for this purpose. The difficulty in the optical modeling stage lies in style diversity of the handwriting scripts. Statistical optical models are traditionally used to tackle this problem such as Hidden Markov models (HMM-GMM) and more recently recurrent neural networks (BLSTM-CTC). Using BLSTM we achieve state of the art performance on the RIMES (for French) and IAM (for English) datasets. The language modeling stage implies the integration of a lexicon and a statistical language model to the recognition processing chain in order to constrain the recognition hypotheses to the most probable sequence of words (sentence) from the language point of view. The difficulty at this stage is related to the finding the optimal vocabulary with minimum Out-Of-Vocabulary words rate (OOV). Enhanced language modeling approaches has been introduced by using sub-lexical units made of syllables or multigrams. The sub-lexical units cover an important portion of the OOV words. Then the language coverage depends on the domain of the language model training corpus, thus the need to train the language model with in domain data. The recognition system performance with the sub-lexical units outperformes the traditional recognition systems that use words or characters language models, in case of high OOV rates. Otherwise equivalent performances are obtained with a compact sub-lexical language model. Thanks to the compact lexicon size of the sub-lexical units, a unified multilingual recognition system has been designed. The unified system performance have been evaluated on the RIMES and IAM datasets. The unified multilingual system shows enhanced recognition performance over the specialized systems, especially when a unified optical model is used
Eickenberg, Michael. "Évaluation de modèles computationnels de la vision humaine en imagerie par résonance magnétique fonctionnelle". Thesis, Paris 11, 2015. http://www.theses.fr/2015PA112206/document.
Pełny tekst źródłaBlood-oxygen-level dependent (BOLD) functional magnetic resonance imaging (fMRI) makes it possible to measure brain activity through blood flow to areas with metabolically active neurons. In this thesis we use these measurements to evaluate the capacity of biologically inspired models of vision coming from computer vision to represent image content in a similar way as the human brain. The main vision models used are convolutional networks.Deep neural networks have made unprecedented progress in many fields in recent years. Even strongholds of biological systems such as scene analysis and object detection have been addressed with enormous success. A body of prior work has been able to establish firm links between the first and last layers of deep convolutional nets and brain regions: The first layer and V1 essentially perform edge detection and the last layer as well as inferotemporal cortex permit a linear read-out of object category. In this work we have generalized this correspondence to all intermediate layers of a convolutional net. We found that each layer of a convnet maps to a stage of processing along the ventral stream, following the hierarchy of biological processing: Along the ventral stream we observe a stage-by-stage increase in complexity. Between edge detection and object detection, for the first time we are given a toolbox to study the intermediate processing steps.A preliminary result to this was obtained by studying the response of the visual areas to presentation of visual textures and analysing it using convolutional scattering networks.The other global aspect of this thesis is “decoding” models: In the preceding part, we predicted brain activity from the stimulus presented (this is called “encoding”). Predicting a stimulus from brain activity is the inverse inference mechanism and can be used as an omnibus test for presence of this information in brain signal. Most often generalized linear models such as linear or logistic regression or SVMs are used for this task, giving access to a coefficient vector the same size as a brain sample, which can thus be visualized as a brain map. However, interpretation of these maps is difficult, because the underlying linear system is either ill-defined and ill-conditioned or non-adequately regularized, resulting in non-informative maps. Supposing a sparse and spatially contiguous organization of coefficient maps, we build on the convex penalty consisting of the sum of total variation (TV) seminorm and L1 norm (“TV+L1”) to develop a penalty grouping an activation term with a spatial derivative. This penalty sets most coefficients to zero but permits free smooth variations in active zones, as opposed to TV+L1 which creates flat active zones. This method improves interpretability of brain maps obtained through cross-validation to determine the best hyperparameter.In the context of encoding and decoding models, we also work on improving data preprocessing in order to obtain the best performance. We study the impulse response of the BOLD signal: the hemodynamic response function. To generate activation maps, instead of using a classical linear model with fixed canonical response function, we use a bilinear model with spatially variable hemodynamic response (but fixed across events). We propose an efficient optimization algorithm and show a gain in predictive capacity for encoding and decoding models on different datasets
Botella, Christophe. "Méthodes statistiques pour la modélisation de la distribution spatiale des espèces végétales à partir de grandes masses d’observations incertaines issues de programmes de sciences citoyennes". Thesis, Montpellier, 2019. http://www.theses.fr/2019MONTS135.
Pełny tekst źródłaHuman botanical expertise is becoming too scarce to provide the field data needed to monitor plant biodiversity. The use of geolocated botanical observations from major citizen science projects, such as Pl@ntNet, opens interesting paths for a temporal monitoring of plant species distribution. Pl@ntNet provides automatically identified flora observations, a confidence score, and can thus be used for species distribution models (SDM). They enable to monitor the distribution of invasive or rare plants, as well as the effects of global changes on species, if we can (i) take into account identification uncertainty, (ii) correct for spatial sampling bias, and (iii) predict species abundances accurately at a fine spatial grain.First, we ask ourselves if we can estimate realistic distributions of invasive plant species on automatically identified occurrences of Pl@ntNet, and what is the effect of filtering with a confidence score threshold. Filtering improves predictions when the confidence level increases until the sample size is limiting. The predicted distributions are generally consistent with expert data, but also indicate urban areas of abundance due to ornamental cultivation and new areas of presence.Next, we studied the correction of spatial sampling bias in SDMs based on presences only. We first mathematically analyzed the bias when the occurrences of a target group of species (Target Group Background, TGB) are used as background points, and compared this bias with that of a spatially uniform selection of base points. We then show that the bias of TGB is due to the variation in the cumulative abundance of target group species in the environmental space, which is difficult to control. We can alternatively jointly model the global observation effort with the abundances of several species. We model the observation effort as a step spatial function defined on a mesh of geographical cells. The addition of massively observed species to the model then reduces the variance in the estimation of the observation effort and thus on the models of the other species.Finally, we propose a new type of SDM based on convolutional neural networks using environmental images as input variables. These models can capture complex spatial patterns of several environmental variables. We propose to share the architecture of the neural network between several species in order to extract common high-level predictors and regularize the model. Our results show that this model outperforms existing SDMs, that performance is improved by simultaneously predicting many species, and this is confirmed by two cooperative SDM evaluation campaigns conducted on independent data sets. This supports the hypothesis that there are common environmental models describing the distribution of many species.Our results support the use of Pl@ntnet occurrences for monitoring plant invasions. Joint modelling of multiple species and observation effort is a promising strategy that transforms the bias problem into a more controllable estimation variance problem. However, the effect of certain factors, such as the level of anthropization, on species abundance is difficult to separate from the effect on observation effort with occurrence data. This can be solved by additional protocolled data collection. The deep learning methods developed show good performance and could be used to deploy spatial species prediction services