Дисертації: "Generative competitive neural network"

1

Гайдук, Ірина Вадимівна. "Вирішення транспортної задачі методами машинного навчання". Master's thesis, КПІ ім. Ігоря Сікорського, 2021. https://ela.kpi.ua/handle/123456789/46504.

Повний текст джерела

Анотація:

Магістерська дисертація: 87 с., 27 рисунків, 24 таблиці, 21 джерело. В роботі розглянута класична задача оптимального транспортування. Проведено дослідження відомих методів її вирішення, їх переваги та недоліки, необхідні умови існування оптимального розв’язку. Окрім цього, був запропонований машинний метод вирішення задачі з побудовою та навчанням моделі на основі генеративної нейронної мережі. В роботі було розглянуто загальні відомості про методи вирішення задачі оптимального транспортування при її незбалансованості та масштабованості. Було виконано аналіз результатів трьох різних типів задач, вирішених методом машинного навчання. Об’єктом дослідження є класична задача оптимального транспортування у трьох різних видах. Предметом дослідження є методи машинного навчання, зокрема генеративна змагальна нейронна мережа.
Master’s thesis: 87 pages, 27 figures, 24 tables, 21 sources. Theme: The classical problem of optimal transportation. The conducted research solves it by known methods, their advantages and disadvantages, the necessary conditions for the existence of an optimal solution. This was a proposed machine method for solving problems with the construction and model of learning based on a generative neural network. The paper considered general information on the method of solving the problem of optimal transportation with its unbalance and scalability. The results of three different types of problems solved by the machine learning method were analyzed. The subject of the study is the classical problem of optimal transportation in three different types. The subject of research is the methods of machine learning, in particular the generative competitive neural network.

Стилі APA, Harvard, Vancouver, ISO та ін.

2

Liu, Mengxin. "Generative Neural Network for Portfolio Optimization." Thesis, Mälardalens högskola, Akademin för utbildning, kultur och kommunikation, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:mdh:diva-53027.

Повний текст джерела

Анотація:

This thesis aims to overcome the drawbacks of traditional portfolio optimization by employing Generative Deep Neural Networks on real stock data. The proposed framework is capable of generating return data that have similar statistical characteristics as the original stock data. The result is acquired using Monte Carlo simulation method and presented in terms of individual risk. This method is tested on real Swedish stock market data. A practical example demonstrates how to optimize a portfolio based on the output of the proposed Generative Adversarial Networks.

Стилі APA, Harvard, Vancouver, ISO та ін.

3

Yamazaki, Hiroyuki Vincent. "On Depth and Complexity of Generative Adversarial Networks." Thesis, KTH, Skolan för datavetenskap och kommunikation (CSC), 2017. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-217293.

Повний текст джерела

Анотація:

Although generative adversarial networks (GANs) have achieved state-of-the-art results in generating realistic look- ing images, they are often parameterized by neural net- works with relatively few learnable weights compared to those that are used for discriminative tasks. We argue that this is suboptimal in a generative setting where data is of- ten entangled in high dimensional space and models are ex- pected to benefit from high expressive power. Additionally, in a generative setting, a model often needs to extrapo- late missing information from low dimensional latent space when generating data samples while in a typical discrimina- tive task, the model only needs to extract lower dimensional features from high dimensional space. We evaluate different architectures for GANs with varying model capacities using shortcut connections in order to study the impacts of the capacity on training stability and sample quality. We show that while training tends to oscillate and not benefit from additional capacity of naively stacked layers, GANs are ca- pable of generating samples with higher quality, specifically for images, samples of higher visual fidelity given proper regularization and careful balancing.
Trots att Generative Adversarial Networks (GAN) har lyckats generera realistiska bilder består de än idag av neurala nätverk som är parametriserade med relativt få tränbara vikter jämfört med neurala nätverk som används för klassificering. Vi tror att en sådan modell är suboptimal vad gäller generering av högdimensionell och komplicerad data och anser att modeller med högre kapaciteter bör ge bättre estimeringar. Dessutom, i en generativ uppgift så förväntas en modell kunna extrapolera information från lägre till högre dimensioner medan i en klassificeringsuppgift så behöver modellen endast att extrahera lågdimensionell information från högdimensionell data. Vi evaluerar ett flertal GAN med varierande kapaciteter genom att använda shortcut connections för att studera hur kapaciteten påverkar träningsstabiliteten, samt kvaliteten av de genererade datapunkterna. Resultaten visar att träningen blir mindre stabil för modeller som fått högre kapaciteter genom naivt tillsatta lager men visar samtidigt att datapunkternas kvaliteter kan öka, specifikt för bilder, bilder med hög visuell fidelitet. Detta åstadkoms med hjälp utav regularisering och noggrann balansering.

Стилі APA, Harvard, Vancouver, ISO та ін.

4

Aftab, Nadeem. "Disocclusion Inpainting using Generative Adversarial Networks." Thesis, Mittuniversitetet, Institutionen för informationssystem och –teknologi, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:miun:diva-40502.

Повний текст джерела

Анотація:

The old methods used for images inpainting of the Depth Image Based Rendering (DIBR) process are inefficient in producing high-quality virtual views from captured data. From the viewpoint of the original image, the generated data’s structure seems less distorted in the virtual view obtained by translation but when then the virtual view involves rotation, gaps and missing spaces become visible in the DIBR generated data. The typical approaches for filling the disocclusion tend to be slow, inefficient, and inaccurate. In this project, a modern technique Generative Adversarial Network (GAN) is used to fill the disocclusion. GAN consists of two or more neural networks that compete against each other and get trained. This study result shows that GAN can inpaint the disocclusion with a consistency of the structure. Additionally, another method (Filling) is used to enhance the quality of GAN and DIBR images. The statistical evaluation of results shows that GAN and filling method enhance the quality of DIBR images.

Стилі APA, Harvard, Vancouver, ISO та ін.

5

Amartur, Sundar C. "Competitive recurrent neural network model for clustering of multispectral data." Case Western Reserve University School of Graduate Studies / OhioLINK, 1995. http://rave.ohiolink.edu/etdc/view?acc_num=case1058445974.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

6

Daley, Jr John. "Generating Synthetic Schematics with Generative Adversarial Networks." Thesis, Högskolan Kristianstad, Fakulteten för naturvetenskap, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:hkr:diva-20901.

Повний текст джерела

Анотація:

This study investigates synthetic schematic generation using conditional generative adversarial networks, specifically the Pix2Pix algorithm was implemented for the experimental phase of the study. With the increase in deep neural network’s capabilities and availability, there is a demand for verbose datasets. This in combination with increased privacy concerns, has led to synthetic data generation utilization. Analysis of synthetic images was completed using a survey. Blueprint images were generated and were successful in passing as genuine images with an accuracy of 40%. This study confirms the ability of generative neural networks ability to produce synthetic blueprint images.

Стилі APA, Harvard, Vancouver, ISO та ін.

7

Ionascu, Beatrice. "Modelling user interaction at scale with deep generative methods." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-239333.

Повний текст джерела

Анотація:

Understanding how users interact with a company's service is essential for data-driven businesses that want to better cater to their users and improve their offering. By using a generative machine learning approach it is possible to model user behaviour and generate new data to simulate or recognize and explain typical usage patterns. In this work we introduce an approach for modelling users' interaction behaviour at scale in a client-service model. We propose a novel representation of multivariate time-series data as time pictures that express temporal correlations through spatial organization. This representation shares two key properties that convolutional networks have been built to exploit and allows us to develop an approach based on deep generative models that use convolutional networks as backbone. In introducing this approach of feature learning for time-series data, we expand the application of convolutional neural networks in the multivariate time-series domain, and specifically user interaction data. We adopt a variational approach inspired by the β-VAE framework in order to learn hidden factors that define different user behaviour patterns. We explore different values for the regularization parameter β and show that it is possible to construct a model that learns a latent representation of identifiable and different user behaviours. We show on real-world data that the model generates realistic samples, that capture the true population-level statistics of the interaction behaviour data, learns different user behaviours, and provides accurate imputations of missing data.
Förståelse för hur användare interagerar med ett företags tjänst är essentiell för data-drivna affärsverksamheter med ambitioner om att bättre tillgodose dess användare och att förbättra deras utbud. Generativ maskininlärning möjliggör modellering av användarbeteende och genererande av ny data i syfte att simulera eller identifiera och förklara typiska användarmönster. I detta arbete introducerar vi ett tillvägagångssätt för storskalig modellering av användarinteraktion i en klientservice-modell. Vi föreslår en ny representation av multivariat tidsseriedata i form av tidsbilder vilka representerar temporala korrelationer via spatial organisering. Denna representation delar två nyckelegenskaper som faltningsnätverk har utvecklats för att exploatera, vilket tillåter oss att utveckla ett tillvägagångssätt baserat på på djupa generativa modeller som bygger på faltningsnätverk. Genom att introducera detta tillvägagångssätt för tidsseriedata expanderar vi applicering av faltningsnätverk inom domänen för multivariat tidsserie, specifikt för användarinteraktionsdata. Vi använder ett tillvägagångssätt inspirerat av ramverket β-VAE i syfte att lära modellen gömda faktorer som definierar olika användarmönster. Vi utforskar olika värden för regulariseringsparametern β och visar att det är möjligt att konstruera en modell som lär sig en latent representation av identifierbara och multipla användarbeteenden. Vi visar med verklig data att modellen genererar realistiska exempel vilka i sin tur fångar statistiken på populationsnivå hos användarinteraktionsdatan, samt lär olika användarbeteenden och bidrar med precisa imputationer av saknad data.

Стилі APA, Harvard, Vancouver, ISO та ін.

8

Pagliarini, Silvia. "Modeling the neural network responsible for song learning." Thesis, Bordeaux, 2021. http://www.theses.fr/2021BORD0107.

Повний текст джерела

Анотація:

Pendant la première période de leur vie, les bébés et les jeunes oiseaux présentent des phases de développement vocal comparables : ils écoutent d'abord leurs parents/tuteurs afin de construire une représentation neurale du stimulus auditif perçu, puis ils commencent à produire des sons qui se rapprochent progressivement du chant de leur tuteur. Cette phase d'apprentissage est appelée la phase sensorimotrice et se caractérise par la présence de babillage. Elle se termine lorsque le chant se cristallise, c'est-à-dire lorsqu'il devient semblable à celui produit par les adultes.Il y a des similitudes entre les voies cérébrales responsables de l'apprentissage sensorimoteur chez l'homme et chez les oiseaux. Dans les deux cas, une voie s’occupe de la production vocale et implique des projections directes des zones auditives vers les zones motrices, et une autre voie s’occupe de l’apprentissage vocal, de l'imitation et de la plasticité.Chez les oiseaux, ces circuits cérébraux sont exclusivement dédiés à l'apprentissage du chant, ce qui en fait un modèle idéal pour explorer les mécanismes neuronaux de l’apprentissage vocal par imitation.Cette thèse vise à construire un modèle de l'apprentissage du chant des oiseaux par imitation. De nombreuses études antérieures ont tenté de mettre en œuvre l'apprentissage par imitation dans des modèles informatiques et partagent une structure commune. Ces modèles comprennent des mécanismes d'apprentissage et, éventuellement, des stratégies d'exploration et d'évaluation.Dans ces modèles, une fonction de contrôle moteur permet la production de sons et une réponse sensorielle modélise soit la façon dont le son est perçu, soit la façon dont il façonne la récompense. Les entrées et les sorties de ces fonctions sont dans plusieurs espaces: l'espace moteur (paramètres moteurs), l'espace sensoriel (sons réels), l'espace perceptif (représentation à faible dimension du son) ou l’espace des objectifs (représentation non perceptive du son cible).Le premier modèle proposé est un modèle théorique inverse basé sur un modèle d'apprentissage vocal simplifié où l'espace sensoriel coïncide avec l'espace moteur (c'est-à-dire qu'il n'y a pas de production sonore). Une telle simplification permet d'étudier comment introduire des hypothèses biologiques (par exemple, une réponse non linéaire) dans un modèle d'apprentissage vocal et quels sont les paramètres qui influencent le plus la puissance de calcul du modèle.Afin de disposer d'un modèle complet (capable de percevoir et de produire des sons), nous avions besoin d'une fonction de contrôle moteur capable de reproduire des sons similaires à des données réelles. Nous avons analysé la capacité de WaveGAN (un réseau de génération) à produire des chants de canari réalistes. Dans ce modèle, l'espace d'entrée devient l'espace latent après l'entraînement et permet la représentation d'un ensemble de données à haute dimension dans une variété à plus basse dimension. Nous avons obtenu des chants de canari réalistes en utilisant seulement trois dimensions pour l'espace latent. Des analyses quantitatives et qualitatives démontrent les capacités d'interpolation du modèle, ce qui suggère que le modèle peut être utilisé comme fonction motrice dans un modèle d'apprentissage vocal.La deuxième version du modèle est un modèle d'apprentissage vocal complet avec une boucle action-perception complète (il comprend l'espace moteur, l'espace sensoriel et l'espace perceptif). La production sonore est réalisée par le générateur GAN obtenu précédemment. Un réseau neuronal récurrent classant les syllabes sert de réponse sensorielle perceptive. La correspondance entre l'espace perceptuel et l'espace moteur est apprise par un modèle inverse. Les résultats préliminaires montrent l'impact du taux d'apprentissage lorsque différentes fonctions de réponse sensorielle sont mises en œuvre
During the first period of their life, babies and juvenile birds show comparable phases of vocal development: first, they listen to their parents/tutors in order to build a neural representation of the experienced auditory stimulus, then they start to produce sound and progressively get closer to reproducing their tutor song. This phase of learning is called the sensorimotor phase and is characterized by the presence of babbling, in babies, and subsong, in birds. It ends when the song crystallizes and becomes similar to the one produced by the adults.It is possible to find analogies between brain pathways responsible for sensorimotor learning in humans and birds: a vocal production pathway involves direct projections from auditory areas to motor neurons, and a vocal learning pathway is responsible for imitation and plasticity. The behavioral studies and the neuroanatomical structure of the vocal control circuit in humans and birds provide the basis for bio-inspired models of vocal learning.In particular, birds have brain circuits exclusively dedicated to song learning, making them an ideal model for exploring the representation of vocal learning by imitation of tutors.This thesis aims to build a vocal learning model underlying song learning in birds. An extensive review of the existing literature is discussed in the thesis: many previous studies have attempted to implement imitative learning in computational models and share a common structure. These learning architectures include the learning mechanisms and, eventually, exploration and evaluation strategies. A motor control function enables sound production and sensory response models either how sound is perceived or how it shapes the reward. The inputs and outputs of these functions lie (1)~in the motor space (motor parameters’ space), (2)~in the sensory space (real sounds) and (3)~either in the perceptual space (a low dimensional representation of the sound) or in the internal representation of goals (a non-perceptual representation of the target sound).The first model proposed in this thesis is a theoretical inverse model based on a simplified vocal learning model where the sensory space coincides with the motor space (i.e., there is no sound production). Such a simplification allows us to investigate how to introduce biological assumptions (e.g. non-linearity response) into a vocal learning model and which parameters influence the computational power of the model the most. The influence of the sharpness of auditory selectivity and the motor dimension are discussed.To have a complete model (which is able to perceive and produce sound), we needed a motor control function capable of reproducing sounds similar to real data (e.g. recordings of adult canaries). We analyzed the capability of WaveGAN (a Generative Adversarial Network) to provide a generator model able to produce realistic canary songs. In this generator model, the input space becomes the latent space after training and allows the representation of a high-dimensional dataset in a lower-dimensional manifold. We obtained realistic canary sounds using only three dimensions for the latent space. Among other results, quantitative and qualitative analyses demonstrate the interpolation abilities of the model, which suggests that the generator model we studied can be used as a motor function in a vocal learning model.The second version of the sensorimotor model is a complete vocal learning model with a full action-perception loop (i.e., it includes motor space, sensory space, and perceptual space). The sound production is performed by the GAN generator previously obtained. A recurrent neural network classifying syllables serves as the perceptual sensory response. Similar to the first model, the mapping between the perceptual space and the motor space is learned via an inverse model. Preliminary results show the influence of the learning rate when different sensory response functions are implemented

Стилі APA, Harvard, Vancouver, ISO та ін.

9

Gustafsson, Alexander, and Jonatan Linberg. "Investigation of generative adversarial network training : The effect of hyperparameters on training time and stability." Thesis, Högskolan i Skövde, Institutionen för informationsteknologi, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:his:diva-19847.

Повний текст джерела

Анотація:

Generative Adversarial Networks (GAN) is a technique used to learn the distribution of some dataset in order to generate similar data. GAN models are notoriously difficult to train, which has caused limited deployment in the industry. The results of this study can be used to accelerate the process of making GANs production ready. An experiment was conducted where multiple GAN models were trained, with the hyperparameters Leaky ReLU alpha, convolutional filters, learning rate and batch size as independent variables. A Mann-Whitney U-test was used to compare the training time and training stability of each model to the others’. Except for the Leaky ReLU alpha, changes to the investigated hyperparameters had a significant effect on the training time and stability. This study is limited to a few hyperparameters and values, a single dataset and few data points, further research in the area could look at the generalisability of the results or investigate more hyperparameters.

Стилі APA, Harvard, Vancouver, ISO та ін.

10

Zheng, Yilin. "Text-Based Speech Video Synthesis from a Single Face Image." The Ohio State University, 2019. http://rave.ohiolink.edu/etdc/view?acc_num=osu1572168353691788.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

11

Ngo, Ho Anh Khoa. "Generative Probabilistic Alignment Models for Words and Subwords : a Systematic Exploration of the Limits and Potentials of Neural Parametrizations." Electronic Thesis or Diss., université Paris-Saclay, 2021. http://www.theses.fr/2021UPASG014.

Повний текст джерела

Анотація:

L'alignement consiste à mettre en correspondance des unités au sein de bitextes, associant un texte en langue source et sa traduction dans une langue cible. L'alignement peut se concevoir à plusieurs niveaux: entre phrases, entre groupes de mots, entre mots, voire à un niveau plus fin lorsque l'une des langues est morphologiquement complexe, ce qui implique d'aligner des fragments de mot (morphèmes). L'alignement peut être envisagé également sur des structures linguistiques plus complexes des arbres ou des graphes. Il s'agit d'une tâche complexe, sous-spécifiée, que les humains réalisent avec difficulté. Son automatisation est un problème exemplaire du traitement des langues, historiquement associé aux premiers modèles de traduction probabilistes. L'arrivée à maturité de nouveaux modèles pour le traitement automatique des langues, reposant sur des représentationts distribuées calculées par des réseaux de neurones permet de reposer la question du calcul de ces alignements. Cette recherche vise donc à concevoir des modèles neuronaux susceptibles d'être appris sans supervision pour dépasser certaines des limitations des modèles d'alignement statistique et améliorer l'état de l'art en matière de précision des alignements automatiques
Alignment consists of establishing a mapping between units in a bitext, combining a text in a source language and its translation in a target language. Alignments can be computed at several levels: between documents, between sentences, between phrases, between words, or even between smaller units end when one of the languages is morphologically complex, which implies to align fragments of words (morphemes). Alignments can also be considered between more complex linguistic structures such as trees or graphs. This is a complex, under-specified task that humans accomplish with difficulty. Its automation is a notoriously difficult problem in natural language processing, historically associated with the first probabilistic word-based translation models. The design of new models for natural language processing, based on distributed representations computed by neural networks, allows us to question and revisit the computation of these alignments. This research project, therefore, aims to comprehensively understand the limitations of existing statistical alignment models and to design neural models that can be learned without supervision to overcome these drawbacks and to improve the state of art in terms of alignment accuracy

Стилі APA, Harvard, Vancouver, ISO та ін.

12

Nilsson, Mårten. "Augmenting High-Dimensional Data with Deep Generative Models." Thesis, KTH, Robotik, perception och lärande, RPL, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-233969.

Повний текст джерела

Анотація:

Data augmentation is a technique that can be performed in various ways to improve the training of discriminative models. The recent developments in deep generative models offer new ways of augmenting existing data sets. In this thesis, a framework for augmenting annotated data sets with deep generative models is proposed together with a method for quantitatively evaluating the quality of the generated data sets. Using this framework, two data sets for pupil localization was generated with different generative models, including both well-established models and a novel model proposed for this purpose. The unique model was shown both qualitatively and quantitatively to generate the best data sets. A set of smaller experiments on standard data sets also revealed cases where this generative model could improve the performance of an existing discriminative model. The results indicate that generative models can be used to augment or replace existing data sets when training discriminative models.
Dataaugmentering är en teknik som kan utföras på flera sätt för att förbättra träningen av diskriminativa modeller. De senaste framgångarna inom djupa generativa modeller har öppnat upp nya sätt att augmentera existerande dataset. I detta arbete har ett ramverk för augmentering av annoterade dataset med hjälp av djupa generativa modeller föreslagits. Utöver detta så har en metod för kvantitativ evaulering av kvaliteten hos genererade data set tagits fram. Med hjälp av detta ramverk har två dataset för pupillokalisering genererats med olika generativa modeller. Både väletablerade modeller och en ny modell utvecklad för detta syfte har testats. Den unika modellen visades både kvalitativt och kvantitativt att den genererade de bästa dataseten. Ett antal mindre experiment på standardiserade dataset visade exempel på fall där denna generativa modell kunde förbättra prestandan hos en existerande diskriminativ modell. Resultaten indikerar att generativa modeller kan användas för att augmentera eller ersätta existerande dataset vid träning av diskriminativa modeller.

Стилі APA, Harvard, Vancouver, ISO та ін.

13

Oskarsson, Joel. "Probabilistic Regression using Conditional Generative Adversarial Networks." Thesis, Linköpings universitet, Statistik och maskininlärning, 2020. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-166637.

Повний текст джерела

Анотація:

Regression is a central problem in statistics and machine learning with applications everywhere in science and technology. In probabilistic regression the relationship between a set of features and a real-valued target variable is modelled as a conditional probability distribution. There are cases where this distribution is very complex and not properly captured by simple approximations, such as assuming a normal distribution. This thesis investigates how conditional Generative Adversarial Networks (GANs) can be used to properly capture more complex conditional distributions. GANs have seen great success in generating complex high-dimensional data, but less work has been done on their use for regression problems. This thesis presents experiments to better understand how conditional GANs can be used in probabilistic regression. Different versions of GANs are extended to the conditional case and evaluated on synthetic and real datasets. It is shown that conditional GANs can learn to estimate a wide range of different distributions and be competitive with existing probabilistic regression models.

Стилі APA, Harvard, Vancouver, ISO та ін.

14

Laifa, Oumeima. "A joint discriminative-generative approach for tumour angiogenesis assessment in computational pathology." Electronic Thesis or Diss., Sorbonne université, 2019. http://www.theses.fr/2019SORUS230.

Повний текст джерела

Анотація:

L’angiogenèse est le processus par lequel de nouveaux vaisseaux sanguins se forment à partir du réseaux préexistant. Au cours de l’angiogenèse tumorale, les cellules tumorales sécrètent des facteurs de croissance qui activent la prolifération et la migration des cellules et stimulent la surproduction du facteur de croissance endothélial vasculaire (VEGF). Le rôle fondamental de l’approvisionnement vasculaire dans la croissance tumorale et le developement des thérapies anticancéreuses rend l’évaluation de l’angiogenèse tumorale, cruciale dans l’évaluation de l’effet des thérapies anti-angiogéniques, en tant que thérapie anticancéreuse prometteuse. Dans cette étude, nous établissons un panel quantitatif et qualitatif pour évaluer les structures des vaisseaux sanguins de la tumeur sur des images de fluorescence non invasives et des images histopathologique sur toute la surface tumorale afin d’identifier les caractéristiques architecturales et les mesures quantitatives souvent associées à la réponse thérapeutique ou prédictive de celle-ci. Nous développons un pipeline formé de Markov Random Field (MFR) et Watershed pour segmenter les vaisseaux sanguins et les composants du micro-environnement tumoral afin d’évaluer quantitativement l’effet du médicament anti-angiogénique Pazopanib sur le système vasculaire tumoral et l’interaction avec le micro-environnement de la tumeur. Le pazopanib, agent anti-angiogénèse, a montré un effet direct sur le système vasculaire du réseau tumoral via les cellules endothéliales. Nos résultats montrent une relation spécifique entre la néovascularisation apoptotique et la densité de noyau dans une tumeur murine traitée par Pazopanib. Une évaluation qualitative des vaisseaux sanguins de la tumeur est réalisée dans la suite de l’étude. Nous avons développé un modèle de réseau de neurone discriminant-générateur basé sur un modele d’apprentissage : réseau de neurones convolutionnels (CNN) et un modèle de connaissance basé sur des règles Marked Point Process (MPP) permettant de segmenter les vaisseaux sanguins sur des images très hétérogènes à l’aide de très peu de données annotées. Nous détaillons l’intuition et la conception du modèle discriminatif-génératif, sa similarité avec les Réseaux antagonistes génératifs (GAN) et nous évaluons ses performances sur des données histopathologiques et synthétiques. Les limites et les perspectives de la méthode sont présentées à la fin de notre étude
Angiogenesis is the process through which new blood vessels are formed from pre-existing ones. During angiogenesis, tumour cells secrete growth factors that activate the proliferation and migration of endothelial cells and stimulate over production of the vascular endothelial growth factor (VEGF). The fundamental role of vascular supply in tumour growth and anti-cancer therapies makes the evaluation of angiogenesis crucial in assessing the effect of anti-angiogenic therapies as a promising anti-cancer therapy. In this study, we establish a quantitative and qualitative panel to evaluate tumour blood vessels structures on non-invasive fluorescence images and histopathological slide across the full tumour to identify architectural features and quantitative measurements that are often associated with prediction of therapeutic response. We develop a Markov Random Field (MFRs) and Watershed framework to segment blood vessel structures and tumour micro-enviroment components to assess quantitatively the effect of the anti-angiogenic drug Pazopanib on the tumour vasculature and the tumour micro-enviroment interaction. The anti-angiogenesis agent Pazopanib was showing a direct effect on tumour network vasculature via the endothelial cells crossing the whole tumour. Our results show a specific relationship between apoptotic neovascularization and nucleus density in murine tumor treated by Pazopanib. Then, qualitative evaluation of tumour blood vessels structures is performed in whole slide images, known to be very heterogeneous. We develop a discriminative-generative neural network model based on both learning driven model convolutional neural network (CNN), and rule-based knowledge model Marked Point Process (MPP) to segment blood vessels in very heterogeneous images using very few annotated data comparing to the state of the art. We detail the intuition and the design behind the discriminative-generative model, and we analyze its similarity with Generative Adversarial Network (GAN). Finally, we evaluate the performance of the proposed model on histopathology slide and synthetic data. The limits of this promising framework as its perspectives are shown

Стилі APA, Harvard, Vancouver, ISO та ін.

15

Käll, Viktor, and Erik Piscator. "Particle Filter Bridge Interpolation in GANs." Thesis, KTH, Matematisk statistik, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-301733.

Повний текст джерела

Анотація:

Generative adversarial networks (GANs), a type of generative modeling framework, has received much attention in the past few years since they were discovered for their capacity to recover complex high-dimensional data distributions. These provide a compressed representation of the data where all but the essential features of a sample is extracted, subsequently inducing a similarity measure on the space of data. This similarity measure gives rise to the possibility of interpolating in the data which has been done successfully in the past. Herein we propose a new stochastic interpolation method for GANs where the interpolation is forced to adhere to the data distribution by implementing a sequential Monte Carlo algorithm for data sampling. The results show that the new method outperforms previously known interpolation methods for the data set LINES; compared to the results of other interpolation methods there was a significant improvement measured through quantitative and qualitative evaluations. The developed interpolation method has met its expectations and shown promise, however it needs to be tested on a more complex data set in order to verify that it also scales well.
Generative adversarial networks (GANs) är ett slags generativ modell som har fått mycket uppmärksamhet de senaste åren sedan de upptäcktes för sin potential att återskapa komplexa högdimensionella datafördelningar. Dessa förser en komprimerad representation av datan där enbart de karaktäriserande egenskaperna är bevarade, vilket följdaktligen inducerar ett avståndsmått på datarummet. Detta avståndsmått möjliggör interpolering inom datan vilket har åstadkommits med framgång tidigare. Häri föreslår vi en ny stokastisk interpoleringsmetod för GANs där interpolationen tvingas följa datafördelningen genom att implementera en sekventiell Monte Carlo algoritm för dragning av datapunkter. Resultaten för studien visar att metoden ger bättre interpolationer för datamängden LINES som användes; jämfört med resultaten av tidigare kända interpolationsmetoder syntes en märkbar förbättring genom kvalitativa och kvantitativa utvärderingar. Den framtagna interpolationsmetoden har alltså mött förväntningarna och är lovande, emellertid fordras att den testas på en mer komplex datamängd för att bekräfta att den fungerar väl även under mer generella förhållanden.

Стилі APA, Harvard, Vancouver, ISO та ін.

16

Barri, Alessandro. "Network mechanisms of memory storage in the balanced cortex." Thesis, Paris 5, 2014. http://www.theses.fr/2014PA05T060/document.

Повний текст джерела

Анотація:

Pas de résumé en français
It is generally maintained that one of cortex’ functions is the storage of a large number of memories. In this picture, the physical substrate of memories is thought to be realised in pattern and strengths of synaptic connections among cortical neurons. Memory recall is associated with neuronal activity that is shaped by this connectivity. In this framework, active memories are represented by attractors in the space of neural activity. Electrical activity in cortical neurones in vivo exhibits prominent temporal irregularity. A standard way to account for this phenomenon is to postulate that recurrent synaptic excitation and inhibition as well as external inputs are balanced. In the common view, however, these balanced networks do not easily support the coexistence of multiple attractors. This is problematic in view of memory function. Recently, theoretical studies showed that balanced networks with synapses that exhibit short-term plasticity (STP) are able to maintain multiple stable states. In order to investigate whether experimentally obtained synaptic parameters are consistent with model predictions, we developed a new methodology that is capable to quantify both response variability and STP at the same synapse in an integrated and statistically-principled way. This approach yields higher parameter precision than standard procedures and allows for the use of more efficient stimulation protocols. However, the findings with respect to STP parameters do not allow to make conclusive statements about the validity of synaptic theories of balanced working memory. In the second part of this thesis an alternative theory of cortical memory storage is developed. The theory is based on the assumptions that memories are stored in attractor networks, and that memories are not represented by network states differing in their average activity levels, but by micro-states sharing the same global statistics. Different memories differ with respect to their spatial distributions of firing rates. From this the main result is derived: the balanced state is a necessary condition for extensive memory storage. Furthermore, we analytically calculate memory storage capacities of rate neurone networks. Remarkably, it can be shown that crucial properties of neuronal activity and physiology that are consistent with experimental observations are directly predicted by the theory if optimal memory storage capacity is required

Стилі APA, Harvard, Vancouver, ISO та ін.

17

Lýsek, Jiří. "Optimalizace síťového přepínače pomocí neuronové sítě." Master's thesis, Vysoké učení technické v Brně. Fakulta strojního inženýrství, 2009. http://www.nusl.cz/ntk/nusl-228699.

Повний текст джерела

Анотація:

This thesis deals with the problem of priority network switch, the model of which was developed in the C++ language. The traffic optimization task is solved by the use of several artificial neural networks, which are described, compared to each other and then evaluated which of them is more suitable for this task. The result of this work is a model of network switch and a comparison of computational time complexity of solving the optimization problem using the artificial neural network. The thesis was developed in research project MSM 0021630529 Intelligent Systems in Automation.

Стилі APA, Harvard, Vancouver, ISO та ін.

18

Ackerman, Wesley. "Semantic-Driven Unsupervised Image-to-Image Translation for Distinct Image Domains." BYU ScholarsArchive, 2020. https://scholarsarchive.byu.edu/etd/8684.

Повний текст джерела

Анотація:

We expand the scope of image-to-image translation to include more distinct image domains, where the image sets have analogous structures, but may not share object types between them. Semantic-Driven Unsupervised Image-to-Image Translation for Distinct Image Domains (SUNIT) is built to more successfully translate images in this setting, where content from one domain is not found in the other. Our method trains an image translation model by learning encodings for semantic segmentations of images. These segmentations are translated between image domains to learn meaningful mappings between the structures in the two domains. The translated segmentations are then used as the basis for image generation. Beginning image generation with encoded segmentation information helps maintain the original structure of the image. We qualitatively and quantitatively show that SUNIT improves image translation outcomes, especially for image translation tasks where the image domains are very distinct.

Стилі APA, Harvard, Vancouver, ISO та ін.

19

Nord, Sofia. "Multivariate Time Series Data Generation using Generative Adversarial Networks : Generating Realistic Sensor Time Series Data of Vehicles with an Abnormal Behaviour using TimeGAN." Thesis, KTH, Skolan för elektroteknik och datavetenskap (EECS), 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:kth:diva-302644.

Повний текст джерела

Анотація:

Large datasets are a crucial requirement to achieve high performance, accuracy, and generalisation for any machine learning task, such as prediction or anomaly detection, However, it is not uncommon for datasets to be small or imbalanced since gathering data can be difficult, time-consuming, and expensive. In the task of collecting vehicle sensor time series data, in particular when the vehicle has an abnormal behaviour, these struggles are present and may hinder the automotive industry in its development. Synthetic data generation has become a growing interest among researchers in several fields to handle the struggles with data gathering. Among the methods explored for generating data, generative adversarial networks (GANs) have become a popular approach due to their wide application domain and successful performance. This thesis focuses on generating multivariate time series data that are similar to vehicle sensor readings from the air pressures in the brake system of vehicles with an abnormal behaviour, meaning there is a leakage somewhere in the system. A novel GAN architecture called TimeGAN was trained to generate such data and was then evaluated using both qualitative and quantitative evaluation metrics. Two versions of this model were tested and compared. The results obtained proved that both models learnt the distribution and the underlying information within the features of the real data. The goal of the thesis was achieved and can become a foundation for future work in this field.
När man applicerar en modell för att utföra en maskininlärningsuppgift, till exempel att förutsäga utfall eller upptäcka avvikelser, är det viktigt med stora dataset för att uppnå hög prestanda, noggrannhet och generalisering. Det är dock inte ovanligt att dataset är små eller obalanserade eftersom insamling av data kan vara svårt, tidskrävande och dyrt. När man vill samla tidsserier från sensorer på fordon är dessa problem närvarande och de kan hindra bilindustrin i dess utveckling. Generering av syntetisk data har blivit ett växande intresse bland forskare inom flera områden som ett sätt att hantera problemen med datainsamling. Bland de metoder som undersökts för att generera data har generative adversarial networks (GANs) blivit ett populärt tillvägagångssätt i forskningsvärlden på grund av dess breda applikationsdomän och dess framgångsrika resultat. Denna avhandling fokuserar på att generera flerdimensionell tidsseriedata som liknar fordonssensoravläsningar av lufttryck i bromssystemet av fordon med onormalt beteende, vilket innebär att det finns ett läckage i systemet. En ny GAN modell kallad TimeGAN tränades för att genera sådan data och utvärderades sedan både kvalitativt och kvantitativt. Två versioner av denna modell testades och jämfördes. De erhållna resultaten visade att båda modellerna lärde sig distributionen och den underliggande informationen inom de olika signalerna i den verkliga datan. Målet med denna avhandling uppnåddes och kan lägga grunden för framtida arbete inom detta område.

Стилі APA, Harvard, Vancouver, ISO та ін.

20

Olsson, Jonathan. "Detecting Faulty Piles of Wood using Anomaly Detection Techniques." Thesis, Luleå tekniska universitet, Institutionen för system- och rymdteknik, 2021. http://urn.kb.se/resolve?urn=urn:nbn:se:ltu:diva-83061.

Повний текст джерела

Анотація:

The forestry and the sawmill industry have a lot of incoming and outgoing piles of wood. It's important to maintain quality and efficiency. This motivates an examination of whether machine learning- or more specifically, anomaly detection techniques can be implemented and used to detect faulty shipments. This thesis presents and evaluates some computer vision techniques and some deep learning techniques. Deep learning can be divided into groups; supervised, semi-supervised and unsupervised. In this thesis, all three groups were examined and it covers supervised methods such as Convolutional Neural Networks, semi-supervised methods such as a modified Convolutional Autoencoder (CAE) and lastly, an unsupervised technique such as Generative Adversarial Network (GAN) was being tested and evaluated. A version of a GAN model proved to perform best for this thesis in terms of the accuracy of faulty detecting shipments with an accuracy rate of 68.2% and 79.8\% overall, which was satisfactory given the problems that were discovered during the progress of the thesis.

Стилі APA, Harvard, Vancouver, ISO та ін.

21

Shapero, Samuel Andre. "Configurable analog hardware for neuromorphic Bayesian inference and least-squares solutions." Diss., Georgia Institute of Technology, 2013. http://hdl.handle.net/1853/51719.

Повний текст джерела

Анотація:

Sparse approximation is a Bayesian inference program with a wide number of signal processing applications, such as Compressed Sensing recovery used in medical imaging. Previous sparse coding implementations relied on digital algorithms whose power consumption and performance scale poorly with problem size, rendering them unsuitable for portable applications, and a bottleneck in high speed applications. A novel analog architecture, implementing the Locally Competitive Algorithm (LCA), was designed and programmed onto a Field Programmable Analog Arrays (FPAAs), using floating gate transistors to set the analog parameters. A network of 6 coefficients was demonstrated to converge to similar values as a digital sparse approximation algorithm, but with better power and performance scaling. A rate encoded spiking algorithm was then developed, which was shown to converge to similar values as the LCA. A second novel architecture was designed and programmed on an FPAA implementing the spiking version of the LCA with integrate and fire neurons. A network of 18 neurons converged on similar values as a digital sparse approximation algorithm, with even better performance and power efficiency than the non-spiking network. Novel algorithms were created to increase floating gate programming speed by more than two orders of magnitude, and reduce programming error from device mismatch. A new FPAA chip was designed and tested which allowed for rapid interfacing and additional improvements in accuracy. Finally, a neuromorphic chip was designed, containing 400 integrate and fire neurons, and capable of converging on a sparse approximation solution in 10 microseconds, over 1000 times faster than the best digital solution.

Стилі APA, Harvard, Vancouver, ISO та ін.

22

Holub, Jiří. "Zvýšení kvality fotografie s použitím hlubokých neuronových sítí." Master's thesis, Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií, 2018. http://www.nusl.cz/ntk/nusl-377334.

Повний текст джерела

Анотація:

This diploma thesis deals with image super-resolution with conservation of good quality. Firstly, there are described state of the art methods dealing with this problem, as well as principles of neural networks with focus on convolutional ones. Finally, there is described a few models of convolutional neural network for image super-resolution to double size, which have been trained, tested and compared on newly created database with pictures of people.

Стилі APA, Harvard, Vancouver, ISO та ін.

23

Mattos, CÃsar Lincoln Cavalcante. "ComitÃs de Classificadores Baseados nas Redes SOM e Fuzzy ART com Sintonia de ParÃmetros e SeleÃÃo de Atributos via MetaheurÃsticas EvolucionÃrias." Universidade Federal do CearÃ, 2011. http://www.teses.ufc.br/tde_busca/arquivo.php?codArquivo=7034.

Повний текст джерела

Анотація:

CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior
O paradigma de classificaÃÃo baseada em comitÃs tem recebido considerÃvel atenÃÃo na literatura cientÃfica em anos recentes. Neste contexto, redes neurais supervisionadas tÃm sido a escolha mais comum para compor os classificadores base dos comitÃs. Esta dissertaÃÃo tem a intenÃÃo de projetar e avaliar comitÃs de classificadores obtidos atravÃs de modificaÃÃes impostas a algoritmos de aprendizado nÃo-supervisionado, tais como as redes Fuzzy ART e SOM, dando origem, respectivamente, Ãs arquiteturas ARTIE (ART in Ensembles) e MUSCLE (Multiple SOM Classifiers in Ensembles). A sintonia dos parÃmetros e a seleÃÃo dos atributos das redes neurais que compÃem as arquiteturas ARTIE e MUSCLE foram tratados por otimizaÃÃo metaheurÃstica, a partir da proposiÃÃo do algoritmo I-HPSO (Improved Hybrid Particles Swarm Optimization). As arquiteturas ARTIE e MUSCLE foram avaliadas e comparadas com comitÃs baseados nas redes Fuzzy ARTMAP, LVQ e ELM em 12 conjuntos de dados reais. Os resultados obtidos indicam que as arquiteturas propostas apresentam desempenhos superiores aos dos comitÃs baseados em redes neurais supervisionadas.

Стилі APA, Harvard, Vancouver, ISO та ін.

24

Marek, Jan. "Rekonstrukce chybějících části obličeje pomocí neuronové sítě." Master's thesis, Vysoké učení technické v Brně. Fakulta informačních technologií, 2020. http://www.nusl.cz/ntk/nusl-433506.

Повний текст джерела

Анотація:

Cílem této práce je vytvořit neuronovou síť která bude schopna rekonstruovat obličeje z fotografií na kterých je část obličeje překrytá maskou. Jsou prezentovány koncepty využívané při vývoji konvolučních neuronových sítí a generativních kompetitivních sítí. Dále jsou popsány koncepty používané v neuronových sítích specificky pro rekonstrukci fotografií obličejů. Je představen model generativní kompetitivní sítě využívající kombinaci hrazených konvolučních vrstev a víceškálových bloků schopný realisticky doplnit oblasti obličeje zakryté maskou.

Стилі APA, Harvard, Vancouver, ISO та ін.

25

Massaccesi, Luciano. "Machine Learning Software for Automated Satellite Telemetry Monitoring." Bachelor's thesis, Alma Mater Studiorum - Università di Bologna, 2020. http://amslaurea.unibo.it/20502/.

Повний текст джерела

Анотація:

During the lifetime of a satellite malfunctions may occur. Unexpected behaviour are monitored using sensors all over the satellite. The telemetry values are then sent to Earth and analysed seeking for anomalies. These anomalies could be detected by humans, but this is considerably expensive. To lower the costs, machine learning techniques can be applied. In this research many diferent machine learning techniques are tested and compared using satellite telemetry data provided by OHB System AG. The fact that the anomalies are collective, together with some data properties, is exploited to improve the performances of the machine learning algorithms. Since the data comes from a real spacecraft, it presents some defects. The data covers in fact a small time-lapse and does not present critical anomalies due to the spacecraft healthiness. Some steps are then taken to improve the evaluation of the algorithms.

Стилі APA, Harvard, Vancouver, ISO та ін.

26

Bak, Adam. "Simulace projevu kožního onemocnění s využitím GAN." Master's thesis, Vysoké učení technické v Brně. Fakulta informačních technologií, 2021. http://www.nusl.cz/ntk/nusl-445569.

Повний текст джерела

Анотація:

Cieľom tejto diplomovej práce je vygenerovanie datasetu syntetických snímkov odtlačkov prstov, ktoré vykazujú známky kožných ochorení. Práca sa zaoberá poškodením spôsobeným kožnými ochoreniami v odtlačkoch prstov a generovaním syntetických odtlačkov prstov. Odtlačky prstov s prejavom kožných ochorení boli generované s využitím modelu založeného na Wasserstein GAN s penalizáciou gradientu. Na trénovanie GAN modelu bola použitá unikátna databáza odtlačkov prstov s prejavom kožných ochorení vytvorená na FIT VUT. Daný model bol trénovaný na troch typoch kožných ochorení: atopický ekzém, psoriáza a dyshidrotický ekzém. Sieť generátoru z natrénovaného WGAN-GP modelu bola použitá na vygenerovanie datasetov syntetických odtlačkov prstov. Tieto syntetické odtlačky boli porovnané s reálnymi odtlačkami s využitím NFIQ a FiQiVi nástrojov na určenie kvality spoločne s porovnaním rozložení lokácií a orientácii markantov v snímkoch odtlačkov prstov.

Стилі APA, Harvard, Vancouver, ISO та ін.

27

Antipov, Grigory. "Apprentissage profond pour la description sémantique des traits visuels humains." Thesis, Paris, ENST, 2017. http://www.theses.fr/2017ENST0071/document.

Повний текст джерела

Анотація:

Les progrès récents des réseaux de neurones artificiels (plus connus sous le nom d'apprentissage profond) ont permis d'améliorer l’état de l’art dans plusieurs domaines de la vision par ordinateur. Dans cette thèse, nous étudions des techniques d'apprentissage profond dans le cadre de l’analyse du genre et de l’âge à partir du visage humain. En particulier, deux problèmes complémentaires sont considérés : (1) la prédiction du genre et de l’âge, et (2) la synthèse et l’édition du genre et de l’âge.D’abord, nous effectuons une étude détaillée qui permet d’établir une liste de principes pour la conception et l’apprentissage des réseaux de neurones convolutifs (CNNs) pour la classification du genre et l’estimation de l’âge. Ainsi, nous obtenons les CNNs les plus performants de l’état de l’art. De plus, ces modèles nous ont permis de remporter une compétition internationale sur l’estimation de l’âge apparent. Nos meilleurs CNNs obtiennent une précision moyenne de 98.7% pour la classification du genre et une erreur moyenne de 4.26 ans pour l’estimation de l’âge sur un corpus interne particulièrement difficile.Ensuite, afin d’adresser le problème de la synthèse et de l’édition d’images de visages, nous concevons un modèle nommé GA-cGAN : le premier réseau de neurones génératif adversaire (GAN) qui produit des visages synthétiques réalistes avec le genre et l’âge souhaités. Enfin, nous proposons une nouvelle méthode permettant d’employer GA-cGAN pour le changement du genre et de l’âge tout en préservant l’identité dans les images synthétiques. Cette méthode permet d'améliorer la précision d’un logiciel sur étagère de vérification faciale en présence d’écarts d’âges importants
The recent progress in artificial neural networks (rebranded as deep learning) has significantly boosted the state-of-the-art in numerous domains of computer vision. In this PhD study, we explore how deep learning techniques can help in the analysis of gender and age from a human face. In particular, two complementary problem settings are considered: (1) gender/age prediction from given face images, and (2) synthesis and editing of human faces with the required gender/age attributes.Firstly, we conduct a comprehensive study which results in an empirical formulation of a set of principles for optimal design and training of gender recognition and age estimation Convolutional Neural Networks (CNNs). As a result, we obtain the state-of-the-art CNNs for gender/age prediction according to the three most popular benchmarks, and win an international competition on apparent age estimation. On a very challenging internal dataset, our best models reach 98.7% of gender classification accuracy and an average age estimation error of 4.26 years.In order to address the problem of synthesis and editing of human faces, we design and train GA-cGAN, the first Generative Adversarial Network (GAN) which can generate synthetic faces of high visual fidelity within required gender and age categories. Moreover, we propose a novel method which allows employing GA-cGAN for gender swapping and aging/rejuvenation without losing the original identity in synthetic faces. Finally, in order to show the practical interest of the designed face editing method, we apply it to improve the accuracy of an off-the-shelf face verification software in a cross-age evaluation scenario

Стилі APA, Harvard, Vancouver, ISO та ін.

28

Hubený, Marek. "Koncepty strojového učení pro kategorizaci objektů v obrazu." Master's thesis, Vysoké učení technické v Brně. Fakulta elektrotechniky a komunikačních technologií, 2017. http://www.nusl.cz/ntk/nusl-316388.

Повний текст джерела

Анотація:

This work is focused on objects and scenes recognition using machine learning and computer vision tools. Before the solution of this problem has been studied basic phases of the machine learning concept and statistical models with accent on their division into discriminative and generative method. Further, the Bag-of-words method and its modification have been investigated and described. In the practical part of this work, the implementation of the Bag-of-words method with the SVM classifier was created in the Matlab environment and the model was tested on various sets of publicly available images.

Стилі APA, Harvard, Vancouver, ISO та ін.

29

Wei, Wen. "Apprentissage automatique des altérations cérébrales causées par la sclérose en plaques en neuro-imagerie multimodale." Thesis, Université Côte d'Azur, 2020. http://www.theses.fr/2020COAZ4021.

Повний текст джерела

Анотація:

La sclérose en plaques (SEP) est la maladie neurologique évolutive la plus courante chez les jeunes adultes dans le monde et représente donc un problème de santé publique majeur avec environ 90 000 patients en France et plus de 500 000 personnes atteintes de SEP en Europe. Afin d'optimiser les traitements, il est essentiel de pouvoir mesurer et suivre les altérations cérébrales chez les patients atteints de SEP. En fait, la SEP est une maladie aux multiples facettes qui implique différents types d'altérations, telles que les dommages et la réparation de la myéline. Selon cette observation, la neuroimagerie multimodale est nécessaire pour caractériser pleinement la maladie. L'imagerie par résonance magnétique (IRM) est devenue un biomarqueur d'imagerie fondamental pour la sclérose en plaques en raison de sa haute sensibilité à révéler des anomalies tissulaires macroscopiques chez les patients atteints de SEP. L'IRM conventionnelle fournit un moyen direct de détecter les lésions de SEP et leurs changements, et joue un rôle dominant dans les critères diagnostiques de la SEP. De plus, l'imagerie par tomographie par émission de positons (TEP), une autre modalité d'imagerie, peut fournir des informations fonctionnelles et détecter les changements tissulaires cibles au niveau cellulaire et moléculaire en utilisant divers radiotraceurs. Par exemple, en utilisant le radiotraceur [11C]PIB, la TEP permet une mesure pathologique directe de l'altération de la myéline. Cependant, en milieu clinique, toutes les modalités ne sont pas disponibles pour diverses raisons. Dans cette thèse, nous nous concentrons donc sur l'apprentissage et la prédiction des altérations cérébrales dérivées des modalités manquantes dans la SEP à partir de données de neuroimagerie multimodale
Multiple Sclerosis (MS) is the most common progressive neurological disease of young adults worldwide and thus represents a major public health issue with about 90,000 patients in France and more than 500,000 people affected with MS in Europe. In order to optimize treatments, it is essential to be able to measure and track brain alterations in MS patients. In fact, MS is a multi-faceted disease which involves different types of alterations, such as myelin damage and repair. Under this observation, multimodal neuroimaging are needed to fully characterize the disease. Magnetic resonance imaging (MRI) has emerged as a fundamental imaging biomarker for multiple sclerosis because of its high sensitivity to reveal macroscopic tissue abnormalities in patients with MS. Conventional MR scanning provides a direct way to detect MS lesions and their changes, and plays a dominant role in the diagnostic criteria of MS. Moreover, positron emission tomography (PET) imaging, an alternative imaging modality, can provide functional information and detect target tissue changes at the cellular and molecular level by using various radiotracers. For example, by using the radiotracer [11C]PIB, PET allows a direct pathological measure of myelin alteration. However, in clinical settings, not all the modalities are available because of various reasons. In this thesis, we therefore focus on learning and predicting missing-modality-derived brain alterations in MS from multimodal neuroimaging data

Стилі APA, Harvard, Vancouver, ISO та ін.

30

Prokopetc, Kristina. "Precise Mapping for Retinal Photocoagulation in SLIM (Slit-Lamp Image Mosaicing)." Thesis, Université Clermont Auvergne‎ (2017-2020), 2017. http://www.theses.fr/2017CLFAC093/document.

Повний текст джерела

Анотація:

Cette thèse est issue d’un accord CIFRE entre le groupe de recherche EnCoV de l’Université Clermont Auvergne et la société Quantel Medical (www.quantel-medical.fr). Quantel Medical est une entreprise spécialisée dans le développement innovant des ultrasons et des produits laser en ophtalmologie. Cette thèse présente un travail de recherche visant à l’application du diagnostic assisté par ordinateur et du traitement des maladies de la rétine avec une utilisation du prototype industriel TrackScan développé par Quantel Medical. Plus précisément, elle contribue au problème du mosaicing précis de l’image de la lampe à fente (SLIM) et du recalage automatique et multimodal en utilisant les images SLIM avec l’angiographie par fluorescence (FA) pour aider à la photo coagulation pan-rétienne naviguée. Nous abordons trois problèmes différents.Le premier problème est lié à l’accumulation des erreurs du recalage en SLIM., il dérive de la mosaïque. Une approche commune pour obtenir la mosaïque consiste à calculer des transformations uniquement entre les images temporellement consécutives dans une séquence, puis à les combiner pour obtenir la transformation entre les vues non consécutives temporellement. Les nombreux algorithmes existants suivent cette approche. Malgré le faible coût de calcul et la simplicité de cette méthode, en raison de sa nature de ‘chaînage’, les erreurs d’alignement s’accumulent, ce qui entraîne une dérive des images dans la mosaïque. Nous proposons donc d’utilise les récents progrès réalisés dans les méthodes d’ajustement de faisceau et de présenter un cadre de réduction de la dérive spécialement conçu pour SLIM. Nous présentons aussi une nouvelle procédure de raffinement local.Deuxièmement, nous abordons le problème induit par divers types d’artefacts communs á l’imagerie SLIM. Ceus-sont liés à la lumière utilisée, qui dégrade considérablement la qualité géométrique et photométrique de la mosaïque. Les solutions existantes permettent de faire face aux blouissements forts qui corrompent entièrement le rendu de la rétine dans l’image tout en laissant de côté la correction des reflets spéculaires semi-transparents et reflets des lentilles. Cela introduit des images fantômes et des pertes d’information. En outre, les méthodes génériques ne produisent pas de résultats satisfaisants dans SLIM. Par conséquent, nous proposons une meilleure alternative en concevant une méthode basée sur une technique rapide en utilisant une seule image pour éliminer les éblouissements et la notion de feux spéculaires semi-transparents en utilisant les indications de mouvement pour la correction intelligente de reflet de lentille.Finalement, nous résolvons le problème du recalage multimodal automatique avec SLIM. Il existe une quantité importante de travaux sur le recalage multimodal de diverses modalités d’image rétinienne. Cependant, la majorité des méthodes existantes nécessitent une détection de points clés dans les deux modalités d’image, ce qui est une tâche très difficile. Dans le cas de SLIM et FA ils ne tiennent pas compte du recalage précis dans la zone maculaire - le repère prioritaire. En outre, personne n’a développé une solution entièrement automatique pour SLIM et FA. Dans cette thèse, nous proposons la première méthode capable de recolu ces deux modalités sans une saisie manuelle, en détectant les repères anatomiques uniquement sur une seule image pour assurer un recalage précis dans la zone maculaire. (...)
This thesis arises from an agreement Convention Industrielle de Formation par la REcherche (CIFRE) between the Endoscopy and Computer Vision (EnCoV) research group at Université Clermont Auvergne and the company Quantel Medical (www.quantel-medical.fr), which specializes in the development of innovative ultrasound and laser products in ophthalmology. It presents a research work directed at the application of computer-aided diagnosis and treatment of retinal diseases with a use of the TrackScan industrial prototype developed at Quantel Medical. More specifically, it contributes to the problem of precise Slit-Lamp Image Mosaicing (SLIM) and automatic multi-modal registration of SLIM with Fluorescein Angiography (FA) to assist navigated pan-retinal photocoagulation. We address three different problems.The first is a problem of accumulated registration errors in SLIM, namely the mosaicing drift.A common approach to image mosaicking is to compute transformations only between temporally consecutive images in a sequence and then to combine them to obtain the transformation between non-temporally consecutive views. Many existing algorithms follow this approach. Despite the low computational cost and the simplicity of such methods, due to its ‘chaining’ nature, alignment errors tend to accumulate, causing images to drift in the mosaic. We propose to use recent advances in key-frame Bundle Adjustment methods and present a drift reduction framework that is specifically designed for SLIM. We also introduce a new local refinement procedure.Secondly, we tackle the problem of various types of light-related imaging artifacts common in SLIM, which significantly degrade the geometric and photometric quality of the mosaic. Existing solutions manage to deal with strong glares which corrupt the retinal content entirely while leaving aside the correction of semi-transparent specular highlights and lens flare. This introduces ghosting and information loss. Moreover, related generic methods do not produce satisfactory results in SLIM. Therefore, we propose a better alternative by designing a method based on a fast single-image technique to remove glares and the notion of the type of semi-transparent specular highlights and motion cues for intelligent correction of lens flare.Finally, we solve the problem of automatic multi-modal registration of FA and SLIM. There exist a number of related works on multi-modal registration of various retinal image modalities. However, the majority of existing methods require a detection of feature points in both image modalities. This is a very difficult task for SLIM and FA. These methods do not account for the accurate registration in macula area - the priority landmark. Moreover, none has developed a fully automatic solution for SLIM and FA. In this thesis, we propose the first method that is able to register these two modalities without manual input by detecting retinal features only on one image and ensures an accurate registration in the macula area.The description of the extensive experiments that were used to demonstrate the effectiveness of each of the proposed methods is also provided. Our results show that (i) using our new local refinement procedure for drift reduction significantly ameliorates the to drift reduction allowing us to achieve an improvement in precision over the current solution employed in the TrackScan; (ii) the proposed methodology for correction of light-related artifacts exhibits a good efficiency, significantly outperforming related works in SLIM; and (iii) despite our solution for multi-modal registration builds on existing methods, with the various specific modifications made, it is fully automatic, effective and improves the baseline registration method currently used on the TrackScan

Стилі APA, Harvard, Vancouver, ISO та ін.

31

Mamba, Mpendulo, and Mpendulo Mamba. "Automatic Brain Tumor Segmentation with a 3-Dimensional Generative Adversarial Neural Network." Thesis, 2018. http://ndltd.ncl.edu.tw/handle/4m9m82.

Повний текст джерела

Анотація:

碩士
國立臺北科技大學
電資國際專班
106
Brain tumor segmentation is a very crucial task in medical image processing. Early diagnosis of brain tumors plays an important role in improving treatment possibilities and increases the survival rate of the patients. Manual segmentation of the brain tumors for cancer diagnosis, from large amounts of magnetic resonance images (MRI) generated in clinical routine, is a difficult and time consuming task. There is a need for automatic brain image segmentation. In this work, we demonstrate a deep neural network for volumetric segmentation that learns from a series of annotated volumetric images given in the Neuroimaging Informatics Technology Initiative (NIfTI) format. Recently, automatic segmentation using deep learning methods proved effective since these methods achieve state-of-the-art results and can address the problem better than other methods. Deep learning methods can also enable efficient processing and objective evaluation of the large amounts of MRI-based images. We investigate 3D conditional adversarial networks as a novel solution to 3D image segmentation for medical segmentation problems. These networks not only learn the mapping from input images to output images, but also learn a loss function to train the mapping between them. This makes it possible to apply the same generic approach to problems that traditionally would require very different loss formulations. We show that this method is effective at generating slices of segmentation data from 3D labelled maps. We utilize a dataset from the medical image computing and computer assisted intervention (MICCAI), which consists of MRI scans of high-grade gliomas (HGG) which are tumors of the central nervous system and low-grade gliomas (LGG) which are referred to as slow-growing tumors. The proposed model is able to discriminate between well segmented and poorly segmented images and the generative model can create segmentation image masks around the tumors and achieves an 80.57% dice score when compared with the dataset.

Стилі APA, Harvard, Vancouver, ISO та ін.

32

Wu, Meng-Xiu, and 吳孟修. "Image Compression Based on Fuzzy Competitive Learning Neural Network." Thesis, 2003. http://ndltd.ncl.edu.tw/handle/88469142545554927818.

Повний текст джерела

Анотація:

碩士
國立成功大學
工程科學系碩博士班
91
A novel image compression algorithm using fuzzy competitive learning neural network is presented in this thesis. The proposed image compression scheme is based on vector quantization. Then, competitive learning neural network and fuzzy control system are included in this scheme. It modifies the learning rate and scaling function of updating equation, which is used to train the codebook, with competitive learning neural network and fuzzy control system, respectively. In the proposed scheme, mean-square error and rate of mean-square error are the inputs of the fuzzy control system, using the membership function and control rules to design the codebook instantaneously and encode the source image in the meanwhile. The monochrome-CCD camera and image acquisition board of PCI interference are used to demo the proposed scheme. According to the experimental results, our scheme could greatly improve the quality of codebook. And comparing with conventional vector quantization, taking the 1024 4 codebook size for example, about 10 percentage of PSNR (peak signal-to-noise ratio) is increased in experiments.

Стилі APA, Harvard, Vancouver, ISO та ін.

33

Yen, Chang Kun, and 張崑淵. "The Research and Application of Competitive Hopfield Neural Network." Thesis, 2006. http://ndltd.ncl.edu.tw/handle/98064017695098755053.

Повний текст джерела

Анотація:

碩士
大葉大學
電機工程學系碩士在職專班
94
Radar system plays an important role in both defense industries and civil applications. In order to obtain the performance radar system should have good tracking algorithm, therefore, it can obtain high detection probability and reduce the tracking errors. A neural network model is investigated in this thesis. Such a mathematical model applying Hopfield Neural Network to tracking systems will have more accurate tracking results. The proposed tracking procedure is developed in this thesis. Moreover, one simulation program using Matlab is also designed. According to the simulation results, this tracking algorithm have good performance.

Стилі APA, Harvard, Vancouver, ISO та ін.

34

Yang, Shr-shian, and 楊士賢. "Chaotic Competitive Hopfield Neural Network for Medical Image Segmentation." Thesis, 2000. http://ndltd.ncl.edu.tw/handle/91334635333622287045.

Повний текст джерела

Анотація:

碩士
國立雲林科技大學
電子工程與資訊工程技術研究所
88
In this thesis, we use the dynamic of chaotic neural network and map into Hopfield model for medical image segmentation. From chaotic dynamic and Hopfield model structure, we could let wining unit which falls into local stable state by winner-take-all rule escape original local stable state, and increase the probability of falling into global stable state. In the whole structure, we introduce a 2-D chaotic Hopfield neural network and an unsupervised competitive Hopfield neural network to parallel process for medical MR image. In medical image segmentation, we are according to global gray distribution and Hopfield neural network model, so we could let chaotic competitive Hopfield neural network algorithm reach an optimum solution and the optimum solution which is based on Lyapunov energy function. When Lyapunov energy function converges to a stable state, we can get an optimum solution. In algorithm, we proposed two algorithms, 1) According to probability of chaotic dynamic, 2) According to Lyapunov energy function convergence. These two algorithms have different advantages and disadvantages but their algorithm result is better than the existing of algorithms of Hard-C Means, Fuzzy-C Means and CHNN.

Стилі APA, Harvard, Vancouver, ISO та ін.

35

CHUANG, FU-CHEN, and 莊馥甄. "Use the Generative Adversarial Network and Attention Model to customize the Neural Style." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/45nr69.

Повний текст джерела

Анотація:

碩士
東吳大學
數學系
107
With the advancement of the information age, the photo app combined with artificial intelligence in the painting method is more and more popular, and the style conversion characteristics are more and more diversified. In the past,there was little use of abstract art. So the purpose of this research is to create novel abstract perception art images, which is to create a new abstract style. It belongs to the artist's unique art, but at the same time it blends with the pictures of the real world.First, using the creative adversarial networks model to create novel abstract art images. Second, using convolutional neural networks and convolutional block attention module to convert images into high-level features and low-resolution images. And then generated into a high-quality image through the perceptual loss network, producing a perceptual art images, showing the emergence of real-world features, and finally combining these models into a novel abstract perception art images. With this model, it is easier to create your own. .

Стилі APA, Harvard, Vancouver, ISO та ін.

36

Po-JungTsai and 蔡帛融. "Using generative adversarial learning to enhance the attention transfer in convolutional neural network." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/87e4j8.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

37

CHENG, HAO-WEN, and 鄭豪文. "Chip and System Design of Image Enhancement Based on Generative Adversarial Neural Network." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/r2dve4.

Повний текст джерела

Анотація:

碩士
國立臺北科技大學
電子工程系
107
Visual image processing has always been a very important field. With the development of multimedia, we can use images in everywhere. The quality of images is not perfect in our expectation. Therefore, image processing is to carry out the images to be repaired. Not only enhance image of contrast to make clear, but real-time processing is also very important. For example, the vehicle electronic assistant equipment must perform image restoration in a short time, so that the driver can keep safety with enhanced images. Due to the development of deep learning in recent years, we can use the neural network for training and simulating the models. The machine learning method can replace the traditional method, which not only shortens the time but also has higher precision. We use deep learning to perform image decomposition, so that we can get the shadow image and the reflection image, and then do the enhancement for the shadow layer image. That can be faster than the traditional methods and perform better efficient. It is possible to fix the over-exposed or over-dark parts of the image to get more complete information, and the shadow layer repair we also use the neural network by algorithm simulation, the conditional generation can be used to restrict the network. Let the speed of training and testing be more precise and fast. Finally, we make the chip to against the most computational volume of the mathematical formula - convolution, and the overall speed can be improved by the hardware acceleration.

Стилі APA, Harvard, Vancouver, ISO та ін.

38

(11211114), Qingyi Gao. "ADVERSARIAL LEARNING ON ROBUSTNESS AND GENERATIVE MODELS." Thesis, 2021.

Знайти повний текст джерела

Анотація:

In this dissertation, we study two important problems in the area of modern deep learning: adversarial robustness and adversarial generative model. In the first part, we study the generalization performance of deep neural networks (DNNs) in adversarial learning. Recent studies have shown that many machine learning models are vulnerable to adversarial attacks, but much remains unknown concerning its generalization error in this scenario. We focus on the $\ell_\infty$ adversarial attacks produced under the fast gradient sign method (FGSM). We establish a tight bound for the adversarial Rademacher complexity of DNNs based on both spectral norms and ranks of weight matrices. The spectral norm and rank constraints imply that this class of networks can be realized as a subset of the class of a shallow network composed with a low dimensional Lipschitz continuous function. This crucial observation leads to a bound that improves the dependence on the network width compared to previous works and achieves depth independence. We show that adversarial Rademacher complexity is always larger than its natural counterpart, but the effect of adversarial perturbations can be limited under our weight normalization framework.

In the second part, we study deep generative models that receive great success in many fields. It is well-known that the complex data usually does not populate its ambient Euclidean space but resides in a lower-dimensional manifold instead. Thus, misspecifying the latent dimension in generative models will result in a mismatch of latent representations and poor generative qualities. To address these problems, we propose a novel framework called Latent Wasserstein GAN (LWGAN) to fuse the auto-encoder and WGAN such that the intrinsic dimension of data manifold can be adaptively learned by an informative latent distribution. In particular, we show that there exist an encoder network and a generator network in such a way that the intrinsic dimension of the learned encodes distribution is equal to the dimension of the data manifold. Theoretically, we prove the consistency of the estimation for the intrinsic dimension of the data manifold and derive a generalization error bound for LWGAN. Comprehensive empirical experiments verify our framework and show that LWGAN is able to identify the correct intrinsic dimension under several scenarios, and simultaneously generate high-quality synthetic data by samples from the learned latent distribution.

Стилі APA, Harvard, Vancouver, ISO та ін.

39

Susskind, Joshua Matthew. "Interpreting Faces with Neurally Inspired Generative Models." Thesis, 2011. http://hdl.handle.net/1807/29884.

Повний текст джерела

Анотація:

Becoming a face expert takes years of learning and development. Many research programs are devoted to studying face perception, particularly given its prerequisite role in social interaction, yet its fundamental neural operations are poorly understood. One reason is that there are many possible explanations for a change in facial appearance, such as lighting, expression, or identity. Despite general agreement that the brain extracts multiple layers of feature detectors arranged into hierarchies to interpret causes of sensory information, very little work has been done to develop computational models of these processes, especially for complex stimuli like faces. The studies presented in this thesis used nonlinear generative models developed within machine learning to solve several face perception problems. Applying a deep hierarchical neural network, we showed that it is possible to learn representations capable of perceiving facial actions, expressions, and identities, better than similar non-hierarchical architectures. We then demonstrated that a generative architecture can be used to interpret high-level neural activity by synthesizing images in a top-down pass. Using this approach we showed that deep layers of a network can be activated to generate faces corresponding to particular categories. To facilitate training models to learn rich and varied facial features, we introduced a new expression database with the largest number of labeled faces collected to date. We found that a model trained on these images learned to recognize expressions comparably to human observers. Next we considered models trained on pairs of images, making it possible to learn how faces change appearance to take on different expressions. Modeling higher-order associations between images allowed us to efficiently match images of the same type according to a learned pairwise similarity measure. These models performed well on several tasks, including matching expressions and identities, and demonstrated performance superior to competing models. In sum, these studies showed that neural networks that extract highly nonlinear features from images using architectures inspired by the brain can solve difficult face perception tasks with minimal guidance by human experts.

Стилі APA, Harvard, Vancouver, ISO та ін.

40

Peng, Chung-Yun, and 彭中鋆. "A Novel Harmonic Competitive Neural Network─Applied to VQ, Clustering and Classification." Thesis, 1999. http://ndltd.ncl.edu.tw/handle/90209094200523147361.

Повний текст джерела

Анотація:

碩士
國立海洋大學
電機工程學系
87
This thesis presents a harmonic on-line learning algorithm useful for training self-creating and self-organizing competitive neural networks. The resulting network is called Harmonic Competitive Neural Network (HCNN). It is shown that, by employing dual local resource counters to record the activity of each node during the competitive learning process, the equi-error and equi-probable criteria can be coherently harmonized. Training in HCNN is smooth and incremental, it not only achieves the biologically plausible on-line learning property, but it also avoids the stability-and-plasticity dilemma, the dead-node problem, as well as the deficiency of local minimum. Vector quantization, clustering and classification are essential techniques in image processing and pattern recognition. We apply the HCNN to perform the three important tasks. In vector quantization, the proposed HCNN is very effective in on-line learning vector quantization. Comparison studies on learning vector quantization involving stationary and non-stationary, structured and non-structured inputs demonstrate that the HCNN outperforms other competitive networks in terms of quantization error, training speed and harmonization of MSE and entropy. Augmented with an agglomerating algorithm, the HCNN can be easily tailored for clustering tasks. Unlike the k-means algorithm and the MST clustering method, the proposed HCNN-based clustering scheme is fully autonomous in that the number of clusters needs not be given in advance, and it consumes less computation time. Finally, we applied HCNN to learning classification. Tested with the two-spiral and iris data, simulation results have shown that HCNN is capable of performing accurate classification.

Стилі APA, Harvard, Vancouver, ISO та ін.

41

Ruiz, Vito Manuel. "Adaptation in a deep network." Thesis, 2011. http://hdl.handle.net/2152/ETD-UT-2011-05-3156.

Повний текст джерела

Анотація:

Though adaptational effects are found throughout the visual system, the underlying mechanisms and benefits of this phenomenon are not yet known. In this work, the visual system is modeled as a Deep Belief Network, with a novel “post-training” paradigm (i.e. training the network further on certain stimuli) used to simulate adaptation in vivo. An optional sparse variant of the DBN is used to help bring about meaningful and biologically relevant receptive fields, and to examine the effects of sparsification on adaptation in their own right. While results are inconclusive, there is some evidence of an attractive bias effect in the adapting network, whereby the network’s representations are drawn closer to the adapting stimulus. As a similar attractive bias is documented in human perception as a result of adaptation, there is thus evidence that the statistical properties underlying the adapting DBN also have a role in the adapting visual system, including efficient coding and optimal information transfer given limited resources. These results are irrespective of sparsification. As adaptation has never been tested directly in a neural network, to the author’s knowledge, this work sets a precedent for future experiments.
text

Стилі APA, Harvard, Vancouver, ISO та ін.

42

Cheng, Tzu-Kuan, and 鄭子寬. "Neural Mechanism of Attention Network under Competitive Stress Using Simultaneous EEG & fNIRS Recording." Thesis, 2016. http://ndltd.ncl.edu.tw/handle/18117195361184728457.

Повний текст джерела

Анотація:

碩士
國立交通大學
電控工程研究所
104
Understanding the dynamics of brain activity during social interaction is important for understanding our social nature and in turn for improving the life quality of people. However, two important limitations of the existing studies are evident. First, social interaction involves two or more individuals. Second, social interaction in real life occurs in a naturalistic environment. We used EEG/fNIRS simultaneous caps with modified attention network task to explore the hemodynamic and metabolic changes associated with EEG-recorded changes in neuronal activity. There are two aims of this study. First, using attention network task to find the cognitive mechanism of human brain under competitive pressure. Second, we want to know whether different competition results have different phenomenon in two physiological signals. Use data in competition minus in non-competition (normal) and find the influence of competitive pressure. In competitive pressure, ERSP in left frontal has theta band burst before response. Oxygenated hemoglobin (HbO) in competition task is lower and raise slower than normal task. In the comparison of win and lose, ERSP have no different because participants all in a very attention status, thus their ERSP result from event onset to response offset have no significant difference. HbO in win group is higher than lose task. The electrodynamic and hemodynamic signatures of competitive pressure provided in this study might improve the understanding of the neural mechanism of stress.

Стилі APA, Harvard, Vancouver, ISO та ін.

43

Roussakov, Maxime. "Financial time series analysis with competitive neural networks." Thèse, 2017. http://hdl.handle.net/1866/20210.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

44

(8892395), Yao Chen. "Inferential GANs and Deep Feature Selection with Applications." Thesis, 2020.

Знайти повний текст джерела

Анотація:

Deep nueral networks (DNNs) have become popular due to their predictive power and flexibility in model fitting. In unsupervised learning, variational autoencoders (VAEs) and generative adverarial networks (GANs) are two most popular and successful generative models. How to provide a unifying framework combining the best of VAEs and GANs in a principled way is a challenging task. In supervised learning, the demand for high-dimensional data analysis has grown significantly, especially in the applications of social networking, bioinformatics, and neuroscience. How to simultaneously approximate the true underlying nonlinear system and identify relevant features based on high-dimensional data (typically with the sample size smaller than the dimension, a.k.a. small-n-large-p) is another challenging task.

In this dissertation, we have provided satisfactory answers for these two challenges. In addition, we have illustrated some promising applications using modern machine learning methods.

In the first chapter, we introduce a novel inferential Wasserstein GAN (iWGAN) model, which is a principled framework to fuse auto-encoders and WGANs. GANs have been impactful on many problems and applications but suffer from unstable training. The Wasserstein GAN (WGAN) leverages the Wasserstein distance to avoid the caveats in the minmax two-player training of GANs but has other defects such as mode collapse and lack of metric to detect the convergence. The iWGAN model jointly learns an encoder network and a generator network motivated by the iterative primal dual optimization process. The encoder network maps the observed samples to the latent space and the generator network maps the samples from the latent space to the data space. We establish the generalization error bound of iWGANs to theoretically justify the performance of iWGANs. We further provide a rigorous probabilistic interpretation of our model under the framework of maximum likelihood estimation. The iWGAN, with a clear stopping criteria, has many advantages over other autoencoder GANs. The empirical experiments show that the iWGAN greatly mitigates the symptom of mode collapse, speeds up the convergence, and is able to provide a measurement of quality check for each individual sample. We illustrate the ability of iWGANs by obtaining a competitive and stable performance with state-of-the-art for benchmark datasets.

In the second chapter, we present a general framework for high-dimensional nonlinear variable selection using deep neural networks under the framework of supervised learning. The network architecture includes both a selection layer and approximation layers. The problem can be cast as a sparsity-constrained optimization with a sparse parameter in the selection layer and other parameters in the approximation layers. This problem is challenging due to the sparse constraint and the nonconvex optimization. We propose a novel algorithm, called Deep Feature Selection, to estimate both the sparse parameter and the other parameters. Theoretically, we establish the algorithm convergence and the selection consistency when the objective function has a Generalized Stable Restricted Hessian. This result provides theoretical justifications of our method and generalizes known results for high-dimensional linear variable selection. Simulations and real data analysis are conducted to demonstrate the superior performance of our method.

In the third chapter, we develop a novel methodology to classify the electrocardiograms (ECGs) to normal, atrial fibrillation and other cardiac dysrhythmias as defined by the Physionet Challenge 2017. More specifically, we use piecewise linear splines for the feature selection and a gradient boosting algorithm for the classifier. In the algorithm, the ECG waveform is fitted by a piecewise linear spline, and morphological features related to the piecewise linear spline coefficients are extracted. XGBoost is used to classify the morphological coefficients and heart rate variability features. The performance of the algorithm was evaluated by the PhysioNet Challenge database (3658 ECGs classified by experts). Our algorithm achieves an average F1 score of 81% for a 10-fold cross validation and also achieved 81% for F1 score on the independent testing set. This score is similar to the top 9th score (81%) in the official phase of the Physionet Challenge 2017.

In the fourth chapter, we introduce a novel region-selection penalty in the framework of image-on-scalar regression to impose sparsity of pixel values and extract active regions simultaneously. This method helps identify regions of interest (ROI) associated with certain disease, which has a great impact on public health. Our penalty combines the Smoothly Clipped Absolute Deviation (SCAD) regularization, enforcing sparsity, and the SCAD of total variation (TV) regularization, enforcing spatial contiguity, into one group, which segments contiguous spatial regions against zero-valued background. Efficient algorithm is based on the alternative direction method of multipliers (ADMM) which decomposes the non-convex problem into two iterative optimization problems with explicit solutions. Another virtue of the proposed method is that a divide and conquer learning algorithm is developed, thereby allowing scaling to large images. Several examples are presented and the experimental results are compared with other state-of-the-art approaches.

Стилі APA, Harvard, Vancouver, ISO та ін.

45

Honzátko, David. "Využití generativních modelů neuronových sítí v obrazové rekonstrukci." Master's thesis, 2018. http://www.nusl.cz/ntk/nusl-372863.

Повний текст джерела

Анотація:

Recent research in generative models came up with a promising approach to modelling the prior proba- bility of natural images. The architecture of these prior models is based on deep neural networks. Although these priors were primarily designed for generating new natural-like images, its potential use is much broader. One of the possible applications is to use these models for solving the inverse problems in low-level vision (i.e., image reconstruction). This usage is mainly possible because the architecture of these models allows computing the derivative of the prior probability with respect to the input image. The main objective of this thesis is to evaluate the usage of these prior models in image reconstruction. This thesis proposes a novel model-based optimization method to two image reconstruction problems - image denoising and single-image super-resolution (SISR). The proposed method uses optimization algorithms for finding the maximum-a- posteriori probability, which is defined using the above mentioned prior models. The experimental results demonstrate that the proposed approach achieves reconstruction performance competitive with the current state-of-the-art methods, especially regarding SISR.

Стилі APA, Harvard, Vancouver, ISO та ін.

46

Parracho, João Oliveira. "JOINT CODING OF MULTIMODAL BIOMEDICAL IMAGES US ING CONVOLUTIONAL NEURAL NETWORKS." Master's thesis, 2020. http://hdl.handle.net/10400.8/6682.

Повний текст джерела

Анотація:

The massive volume of data generated daily by the gathering of medical images with different modalities might be difficult to store in medical facilities and share through communication networks. To alleviate this issue, efficient compression methods must be implemented to reduce the amount of storage and transmission resources required in such applications. However, since the preservation of all image details is highly important in the medical context, the use of lossless image compression algorithms is of utmost importance. This thesis presents the research results on a lossless compression scheme designed to encode both computerized tomography (CT) and positron emission tomography (PET). Different techniques, such as image-to-image translation, intra prediction, and inter prediction are used. Redundancies between both image modalities are also investigated. To perform the image-to-image translation approach, we resort to lossless compression of the original CT data and apply a cross-modality image translation generative adversarial network to obtain an estimation of the corresponding PET. Two approaches were implemented and evaluated to determine a PET residue that will be compressed along with the original CT. In the first method, the residue resulting from the differences between the original PET and its estimation is encoded, whereas in the second method, the residue is obtained using encoders inter-prediction coding tools. Thus, in alternative to compressing two independent picture modalities, i.e., both images of the original PET-CT pair solely the CT is independently encoded alongside with the PET residue, in the proposed method. Along with the proposed pipeline, a post-processing optimization algorithm that modifies the estimated PET image by altering the contrast and rescaling the image is implemented to maximize the compression efficiency. Four different versions (subsets) of a publicly available PET-CT pair dataset were tested. The first proposed subset was used to demonstrate that the concept developed in this work is capable of surpassing the traditional compression schemes. The obtained results showed gains of up to 8.9% using the HEVC. On the other side, JPEG2k proved not to be the most suitable as it failed to obtain good results, having reached only -9.1% compression gain. For the remaining (more challenging) subsets, the results reveal that the proposed refined post-processing scheme attains, when compared to conventional compression methods, up 6.33% compression gain using HEVC, and 7.78% using VVC.

Стилі APA, Harvard, Vancouver, ISO та ін.

47

Rodrigues, Diogo Manuel de Castro. "Integrating Vision and Language for Automatic Face Descriptions." Master's thesis, 2018. http://hdl.handle.net/10316/86752.

Повний текст джерела

Анотація:

Dissertação de Mestrado Integrado em Engenharia Electrotécnica e de Computadores apresentada à Faculdade de Ciências e Tecnologia
Nesta dissertação, para criar um exemplo único de um sistema de face para texto e texto para face foi integrado visão por computador e processamento de linguagem natural. O propósito é fornecer uma solução que permita ajudar os seres humanos a realizar funções com maior qualidade e de forma mais rápida. Assim sendo pretende-se criar um sistema que possa ser usado, por exemplo, para descrever rostos para pessoas com deficiência visual ou para gerar rostos a partir de descrições para investigações criminais. No entanto trata-se apenas de uma versão preliminar, na medida em que o curto tempo disponível para a sua realização não permitiu alcançar a ambiciosa proposta. De forma a atingir este objectivo, foi criado um sistema com a capacidade de descrever textualmente imagens faciais e por outro lado, gerar automaticamente imagens faciais a partir de descrições textuais. O sistema é dividido em duas partes, a primeira tem como função prever atributos das imagens faciais através de uma rede neuronal convolucional. Estes são utilizados como base para o modelo de geração de linguagem natural, gerando descrições textuais numa metodologia baseada em regras. A segunda parte, usa uma técnica simples de extração de palavras chave para analisar o texto e identificar os atributos nessa descrição. Seguidamente, o sistema usa uma rede generativa adversarial para gerar uma imagem facial com o conjunto das características desejadas. Os atributos são usados como base no nosso método, uma vez que representam um identificador dominante que transmite características sobre um rosto com eficácia.Os resultados demonstraram, mais uma vez, que os métodos CNN e GAN são atualmente as melhores opções para, tarefas de reconhecimento e geração de imagens, respectivamente. Esta conclusão destá assente nos resultados convincentes. Por outro lado, os métodos de processamento de linguagem natural apesar de terem funcionado bem, de acordo com os objectivos, os seus resultados são menos notáveis, especialmente o modelo de geração de linguagem natural. Este trabalho propõe uma solução fiável e funcional para resolver este sistema complexo, no entanto é uma área que merece uma extensa investigação e desenvolvimento.
In this dissertation, computer vision and Natural Language Processing (NLP) are integrated to create a unique example of a face-to-text and text-to-face system. Its intention is to provide a solution that can help humans to perform their jobs with better quality and with a quick response. The aim is to create a system that can be used, for example, to describe faces for visually impaired people or to generate faces from descriptions for criminal investigations. However, this is a preliminary version as it is an ambitious goal to be achieved during the time available for its realization.To accomplish this motivation, a system was created with the capability of describing, textually, facial images, along with the ability to automatically generate face images from text descriptions. The system is divided into two sub-systems. The first part predicts attributes from the face images through a Convolutional Neural Network (CNN) method that are used, further, as a base to the Natural Language Generation (NLG) model. The descriptions are generated on a rule-based methodology. The second part of the system uses a simple keyword extraction technique to analyze the text and identify the attributes on that description. After that, it uses a conditional Generative Adversarial Network (GAN) to generate a facial image with a specific set of desired attributes. The reason why attributes are used as a base on the method is because they are a dominant identifier that can efficiently transmit characteristic about a face. The results demonstrate, once again, that either CNN and GAN methods are presently the best options for recognition and generation tasks, respectively. This conclusion is due to their convincing results. On the other hand, the NLP methods worked well for their purposes. However, its results are less remarkable, especially the NLG model. This work proposes a reliable and functional solution for solving this complex system. Nevertheless, this area needs an extensive investigation and development.

Стилі APA, Harvard, Vancouver, ISO та ін.

48

Dumoulin, Vincent. "Representation Learning for Visual Data." Thèse, 2018. http://hdl.handle.net/1866/21140.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

49

Warde-Farley, David. "Feedforward deep architectures for classification and synthesis." Thèse, 2017. http://hdl.handle.net/1866/20501.

Повний текст джерела

Стилі APA, Harvard, Vancouver, ISO та ін.

50

Santos, Ângelo Emanuel Neves dos. "Design and simulation of a smart bottle with fill-level sensing based on oxide TFT technology." Master's thesis, 2016. http://hdl.handle.net/10362/19593.

Повний текст джерела

Анотація:

Packaging is an important element responsible for brand growth and one of the main rea-sons for producers to gain competitive advantages through technological innovation. In this re-gard, the aim of this work is to design a fully autonomous electronic system for a smart bottle packaging, being integrated in a European project named ROLL-OUT. The desired application for the smart bottle is to act as a fill-level sensor system in order to determine the liquid content level that exists inside an opaque bottle, so the consumer can exactly know the remaining quantity of the product inside. An in-house amorphous indium–gallium–zinc oxide thin-film transistor (a-IGZO TFT) model, previously developed, was used for circuit designing purposes. This model was based in an artificial neural network (ANN) equivalent circuit approach. Taking into account that only n-type oxide TFTs were used, plenty of electronic building-blocks have been designed: clock generator, non-overlapping phase generator, a capacitance-to-voltage converter and a comparator. As it was demonstrated by electrical simulations, it has been achieved good functionality for each block, having a final system with a power dissipation of 2.3 mW (VDD=10 V) not considering the clock generator. Four printed circuit boards (PCBs) have been also designed in order to help in the testing phase. Mask layouts were already designed and are currently in fabrication, foreseeing a suc-cessful circuit fabrication, and a major step towards the design and integration of complex trans-ducer systems using oxide TFTs technology.

Стилі APA, Harvard, Vancouver, ISO та ін.

Дисертації з теми "Generative competitive neural network"

Оформте джерело за APA, MLA, Chicago, Harvard та іншими стилями