Time- series Generative Adversarial Networks (TimeGAN) is proposed to overcome the GAN model’s insufficiency in producing synthetic samples that inherit the predictive ability of the original timeseries data. TimeGAN combines the unsupervised adversarial loss in the GAN framework with a supervised loss adopted from an autoregressive model. However, TimeGAN is like another GANbased model that only learns from the set of smaller sequences extracted from the original time-series. This behavior yields a severe consequence when encountering data augmentation for time-series with multiple seasonal patterns, as found in the mobile telecommunication network data. This study examined the effectiveness of the TimeGAN model with the help of Dynamic Time Warping (DTW) and different types of RNN as its architecture to produce synthetic mobile telecommunication network data, which can be utilized to improve the forecasting performance of the statistical and deep learning models relative to the baseline models trained only on the original data. The experiment results indicate that DTW helps TimeGAN maintaining the multiple seasonal attributes. In addition, either LSTM or Bidirectional LSTM as TimeGAN architecture ensures the model is robust to mode collapse problem and creates synthetic data that are diversified and indistinguishable from the original time-series. Finally, merging both original and synthetic time-series becomes a compelling way to significantly improve the deep learning model’s forecasting performance but fails to do so for the statistical model. / Time-series Generative Adversarial Networks (TimeGAN) föreslås för att övervinna GAN-modellens brist att kunna producera syntetisk data som ärver de prediktiva förmåga från den ursprungliga tidsseriedatan. TimeGAN kombinerar den icke-vägledande förlusten i GAN-ramverket tillsammans med den vägledande förlusten från en autoregressiv modell. TimeGAN liknar en vanlig GAN-baserad modell, men behöver bara en mindre uppsättning sekvenser från den ursprungliga tidsserien för att lära sig. Denna egenskap kan dock leda till allvarliga konsekvenser när man stöter på dataförstoring för tidsserier med flera säsongsmönster, vilket återfinns i mobilnätverksdata. Denna studie har undersökt effektiviteten av TimeGAN-modellen med hjälp av Dynamic Time Warping (DTW) och olika typer av RNN som dess arkitektur för att producera syntetisk mobilnätverksdata. Detta kan användas för att förbättra statistiska och djupinlärningsmodellers prognostisering relativt till modeller som bara har tränat på orginaldata. De experimentella resultaten indikerar att DTW hjälper TimeGAN att bibehålla de olika säsongsattributen. Dessutom, TimeGAN med antingen LSTM eller Bidirectional LSTM som arkitektur säkerställer att modellen är robust för lägesfallsproblem och skapar syntetisk data som är diversifierade och inte kan urskiljas från den ursprungliga tidsserien. Slutligen, en sammanslagning av både ursprungliga och syntetiska tidsserier blir ett övertygande sätt att avsevärt förbättra djupinlärningsmodellens prestanda men misslyckas med detta för den statistiska modellen.

Cette thèse explore deux sujets liés dans le contexte de l'apprentissage profond : l'apprentissage incrémental et la génération des images. L'apprentissage incrémental étudie l'entrainement des modèles dont la fonction objective évolue avec le temps (exemple : Ajout de nouvelles catégories à une tâche de classification). La génération d'images cherche à apprendre une distribution d'images naturelles pour générer de nouvelles images ressemblant aux images de départ.L’apprentissage incrémental est un problème difficile dû au phénomène appelé l'oubli catastrophique : tout changement important de l’objectif au cours de l'entrainement provoque une grave dégradation des connaissances acquises précédemment. Nous présentons un cadre d'apprentissage permettant d'introduire de nouvelles classes dans un réseau de détection d'objets. Il est basé sur l’idée de la distillation du savoir pour lutter les effets de l’oubli catastrophique : une copie fixe du réseau évalue les anciens échantillons et sa sortie est réutilisée dans un objectif auxiliaire pour stabiliser l’apprentissage de nouvelles classes. Notre framework extrait ces échantillons d'anciennes classes à la volée à partir d'images entrantes, contrairement à d'autres solutions qui gardent un sous-ensemble d'échantillons en mémoire.Pour la génération d’images, nous nous appuyons sur le modèle du réseau adverse génératif (en anglais generative adversarial network ou GAN). Récemment, les GANs ont considérablement amélioré la qualité des images générées. Cependant, ils offrent une pauvre couverture de l'ensemble des données : alors que les échantillons individuels sont de grande qualité, certains modes de la distribution d'origine peuvent ne pas être capturés. De plus, contrairement à la mesure de vraisemblance couramment utilisée pour les modèles génératives, les méthodes existantes d'évaluation GAN sont axées sur la qualité de l'image et n'évaluent donc pas la qualité de la couverture du jeu de données. Nous présentons deux approches pour résoudre ces problèmes.La première approche évalue les GANs conditionnels à la classe en utilisant deux mesures complémentaires basées sur la classification d'image - GAN-train et GAN-test, qui approchent respectivement le rappel (diversité) et la précision (qualité d'image) des GANs. Nous évaluons plusieurs approches GANs récentes en fonction de ces deux mesures et démontrons une différence de performance importante. De plus, nous observons que la difficulté croissante du jeu de données, de CIFAR10 à ImageNet, indique une corrélation inverse avec la qualité des GANs, comme le montre clairement nos mesures.Inspirés par notre étude des modèles GANs, la seconde approche applique explicitement la couverture d'un jeux de données pendant la phase d'entrainement de GAN. Nous développons un modèle génératif combinant la qualité d'image GAN et l'architecture VAE dans l'espace latente engendré par un modèle basé sur le flux, Real-NVP. Cela nous permet d’évaluer une vraisemblance correcte et d’assouplir simultanément l’hypothèse d’indépendance dans l’espace RVB qui est courante pour les VAE. Nous obtenons le score Inception et la FID en concurrence avec les GANs à la pointe de la technologie, tout en maintenant une bonne vraisemblance pour cette classe de modèles. / This dissertation explores two related topics in the context of deep learning: incremental learning and image generation. Incremental learning studies training of models with the objective function evolving over time, eg, addition of new categories to a classification task. Image generation seeks to learn a distribution of natural images for generating new images resembling original ones.Incremental learning is a challenging problem due to the phenomenon called catastrophic forgetting: any significant change to the objective during training causes a severe degradation of previously learned knowledge. We present a learning framework to introduce new classes to an object detection network. It is based on the idea of knowledge distillation to counteract catastrophic forgetting effects: fixed copy of the network evaluates old samples and its output is reused in an auxiliary loss to stabilize learning of new classes. Our framework mines these samples of old classes on the fly from incoming images, in contrast to other solutions that keep a subset of samples in memory.On the second topic of image generation, we build on the Generative Adversarial Network (GAN) model. Recently, GANs significantly improved the quality of generated images. However, they suffer from poor coverage of the dataset: while individual samples have great quality, some modes of the original distribution may not be captured. In addition, existing GAN evaluation methods are focused on image quality, and thus do not evaluate how well the dataset is covered, in contrast to the likelihood measure commonly used for generative models. We present two approaches to address these problems.The first method evaluates class-conditional GANs using two complementary measures based on image classification - GAN-train and GAN-test, which approximate recall (diversity) and precision (quality of the image) of GANs respectively. We evaluate several recent GAN approaches based on these two measures, and demonstrate a clear difference in performance. Furthermore, we observe that the increasing difficulty of the dataset, from CIFAR10 over CIFAR100 to ImageNet, shows an inverse correlation with the quality of the GANs, as clearly evident from our measures.Inspired by our study of GAN models, we present a method to explicitly enforce dataset coverage during the GAN training phase. We develop a generative model that combines GAN image quality with VAE architecture in the feature space engendered by a flow-based model Real-NVP. This allows us to evaluate a valid likelihood and simultaneously relax the independence assumption in RGB space which is common for VAEs. We achieve Inception score and FID competitive with state-of-the-art GANs, while maintaining good likelihood for this class of models.

The use of machine learning algorithms to enhance and facilitate medical diagnosis and analysis is a promising and an important area, which could improve the workload of clinicians’ substantially. In order for machine learning algorithms to learn a certain task, large amount of data needs to be available. Data sets for medical image analysis are rarely public due to restrictions concerning the sharing of patient data. The production of synthetic images could act as an anonymization tool to enable the distribution of medical images and facilitate the training of machine learning algorithms, which could be used in practice. This thesis investigates the use of Generative Adversarial Networks (GAN) for synthesis of new thoracic computer tomography (CT) images, with no connection to real patients. It also examines the usefulness of the images by comparing the quantitative performance of a segmentation network trained with the synthetic images with the quantitative performance of the same segmentation network trained with real thoracic CT images. The synthetic thoracic CT images were generated using CycleGAN for image-to-image translation between label map ground truth images and thoracic CT images. The synthetic images were evaluated using different set-ups of synthetic and real images for training the segmentation network. All set-ups were evaluated according to sensitivity, accuracy, Dice and F2-score and compared to the same parameters evaluated from a segmentation network trained with 344 real images. The thesis shows that it was possible to generate synthetic thoracic CT images using GAN. However, it was not possible to achieve an equal quantitative performance of a segmentation network trained with synthetic data compared to a segmentation network trained with the same amount of real images in the scope of this thesis. It was possible to achieve equal quantitative performance of a segmentation network, as a segmentation network trained on real images, by training it with a combination of real and synthetic images, where a majority of the images were synthetic images and a minority were real images. By using a combination of 59 real images and 590 synthetic images, equal performance as a segmentation network trained with 344 real images was achieved regarding sensitivity, Dice and F2-score. Equal quantitative performance of a segmentation network could thus be achieved by using fewer real images together with an abundance of synthetic images, created at close to no cost, indicating a usefulness of synthetically generated images.

This Thesis surveys the landscape of Data Augmentation for image datasets. Completing this survey inspired further study into a method of generative modeling known as Generative Adversarial Networks (GANs). A survey on GANs was conducted to understood recent developments and the problems related to training them. Following this survey, four experiments were proposed to test the application of GANs for data augmentation and to contribute to the quality improvement in GAN-generated data. Experimental results demonstrate the effectiveness of GAN-generated data as a pre-training metric. The other experiments discuss important characteristics of GAN models such as the refining of prior information, transferring generative models from large datasets to small data, and automating the design of Deep Neural Networks within the context of the GAN framework. This Thesis will provide readers with a complete introduction to Data Augmentation and Generative Adversarial Networks, as well as insights into the future of these techniques. / Includes bibliography. / Thesis (M.S.)--Florida Atlantic University, 2019. / FAU Electronic Theses and Dissertations Collection

Environment maps are widely used in several computer graphics fields, such as realistic architectural rendering or computer games as sources of the light in the scene. Obtaining these maps is not easy, since they have to have both a high- dynamic range as well as a high resolution. As a result, they are expensive to make and the supply is limited. Deep neural networks are a widely unexplored research area and have been successfully used for generating complex and realistic images like human portraits. Neural networks perform well at predicting data from complex models, which are easily observable, such as photos of the real world. This thesis explores the idea of generating physically plausible environment maps by utilizing deep neural networks known as generative adversarial networks. Since a skydome dataset is not publicly available, we develop a scalable capture process with both low-end and high-end hardware. We implement a pipeline to process the captured data before feeding it to a network and extend an already existing network architecture to generate HDR environment maps. We then run a series of experiments to determine the quality of the results and uncover the directions of possible further research.

This thesis investigates how synthetic data can be utilized when training convolutional neural networks to detect flags with threatening symbols. The synthetic data used in this thesis consisted of rendered 3D flags with different textures and flags cut out from real images. The synthetic data showed that it can achieve an accuracy above 80% compared to 88% accuracy achieved by a data set containing only real images. The highest accuracy scored was achieved by combining real and synthetic data showing that synthetic data can be used as a complement to real data. Some attempts to improve the accuracy score was made using generative adversarial networks without achieving any encouraging results.

abstract: In the past few decades, there has been a remarkable shift in the boundary between public and private information. The application of information technology and electronic communications allow service providers (businesses) to collect a large amount of data. However, this ``data collection" process can put the privacy of users at risk and also lead to user reluctance in accepting services or sharing data. This dissertation first investigates privacy sensitive consumer-retailers/service providers interactions under different scenarios, and then focuses on a unified framework for various information-theoretic privacy and privacy mechanisms that can be learned directly from data. Existing approaches such as differential privacy or information-theoretic privacy try to quantify privacy risk but do not capture the subjective experience and heterogeneous expression of privacy-sensitivity. The first part of this dissertation introduces models to study consumer-retailer interaction problems and to better understand how retailers/service providers can balance their revenue objectives while being sensitive to user privacy concerns. This dissertation considers the following three scenarios: (i) the consumer-retailer interaction via personalized advertisements; (ii) incentive mechanisms that electrical utility providers need to offer for privacy sensitive consumers with alternative energy sources; (iii) the market viability of offering privacy guaranteed free online services. We use game-theoretic models to capture the behaviors of both consumers and retailers, and provide insights for retailers to maximize their profits when interacting with privacy sensitive consumers. Preserving the utility of published datasets while simultaneously providing provable privacy guarantees is a well-known challenge. In the second part, a novel context-aware privacy framework called generative adversarial privacy (GAP) is introduced. Inspired by recent advancements in generative adversarial networks, GAP allows the data holder to learn the privatization mechanism directly from the data. Under GAP, finding the optimal privacy mechanism is formulated as a constrained minimax game between a privatizer and an adversary. For appropriately chosen adversarial loss functions, GAP provides privacy guarantees against strong information-theoretic adversaries. Both synthetic and real-world datasets are used to show that GAP can greatly reduce the adversary's capability of inferring private information at a small cost of distorting the data. / Dissertation/Thesis / Doctoral Dissertation Electrical Engineering 2018

<div><div><div><p>The presence of attacks in day-to-day traffic flow in connected networks is considerably less compared to genuine traffic flow. Yet, the consequences of these attacks are disastrous. It is very important to identify if the network is being attacked and block these attempts to protect the network system. Failure to block these attacks can lead to loss of confidential information and reputation and can also lead to financial loss. One of the strategies to identify these attacks is to use machine learning algorithms that learn to identify attacks by looking at previous examples. But since the number of attacks is small, it is difficult to train these machine learning algorithms. This study aims to use generative techniques to create new attack samples that can be used to train the machine learning based intrusion detection systems to identify more attacks. Two metrics are used to verify that the training has improved and a binary classifier is used to perform a two-sample test for verifying the generated attacks.</p></div></div></div>

Abundance of labelled data played a crucial role in the recent developments in computer vision, but that faces problems like scalability and transferability to the wild. One alternative approach is to utilize the data without labels, i.e. unsupervised learning, in learning valuable information and put it in use to tackle vision problems. Generative Adversarial Networks (GANs) have gained momentum for their ability to model image distributions in unsupervised manner. They learn to emulate the training set and that enables sampling from that domain and using the knowledge learned for useful applications. Several methods proposed enhancing GANs, including regularizing the loss with some feature matching. We seek to push GANs beyond the data in the training and try to explore unseen territory in the image manifold. We first propose a new regularizer for GAN based on K-Nearest Neighbor (K-NN) selective feature matching to a target set Y in high-level feature space, during the adversarial training of GAN on the base set X, and we call this novel model K-GAN. We show that minimizing the added term follows from cross-entropy minimization between the distributions of GAN and set Y. Then, we introduce a cascaded framework for GANs that try to address the task of imagining a new distribution that combines the base set X and target set Y by cascading sampling GANs with translation GANs, and we dub the cascade of such GANs as the Imaginative Adversarial Network (IAN). Several cascades are trained on a collected dataset Zoo-Faces and generated innovative samples are shown, including from K-GAN cascade. We conduct an objective and subjective evaluation for different IAN setups in the addressed task of generating innovative samples and we show the effect of regularizing GAN on different scores. We conclude with some useful applications for these IANs, like multi-domain manifold traversing.

The study and the analysis of marine ecosystems is a significant part of the marine science research. These systems are valuable resources for fisheries, improving water quality and can even be used in drugs production. The investigation of ichthyoplankton inhabiting these ecosystems is also an important research field. Ichthyoplankton are fish in their early stages of life. In this stage, the fish have relatively similar shape and are small in size. The currently used way of identifying them is not optimal. Marine scientists typically study such organisms by sending a team that collects samples from the sea which is then taken to the lab for further investigation. These samples need to be studied by an expert and usually end needing a DNA sequencing. This method is time-consuming and requires a high level of experience. The recent advances in AI have helped to solve and automate several difficult tasks which motivated us to develop a classification tool for ichthyoplankton. We show that using machine learning techniques, such as generative adversarial networks combined with transfer learning solves such a problem with high accuracy. We show that using traditional machine learning algorithms fails to solve it. We also give a general framework for creating a classification tool when the dataset used for training is a limited dataset. We aim to build a user-friendly tool that can be used by any user for the classification task and we aim to give a guide to the researchers so that they can follow in creating a classification tool.