В диссертации изучается влияние аугментации данных скользящим окном на производительность различных архитектур рекуррентных нейронных сетей (RNN) для классификации временных рядов. Исследование оценивает модели на основе слоев долговременной краткосрочной памяти (LSTM), SimpleRNN, Gated Recurrent Unit (GRU) и гибридной RNN, применяемые к классификации пяти видов деятельности: езда на велосипеде, катание на роликовых лыжах (R-Skiing), бег, катание на лыжах и ходьба. Результаты показывают, что аугментация данных скользящим окном значительно повышает производительность модели, улучшая ключевые показатели, такие как точность, отзыв, F1-оценка и достоверность. Среди протестированных моделей модели гибридной RNN и GRU продемонстрировали наивысшую точность и возможности обобщения. Кроме того, мы протестировали несколько размеров окна и шага. Конфигурация с большим размером окна (256) в целом дала лучшие результаты. Эти результаты согласуются с существующей литературой, подчеркивая эффективность аугментации данных и передовых архитектур RNN в классификации временных рядов. Исследование подчеркивает важность дополнения данных для повышения надежности моделей и предоставляет ценную информацию для будущих исследований и практических приложений в различных областях. / The thesis explores the impact of sliding window data augmentation on the performance of various Recurrent Neural Network (RNN) architectures for time series classification. The study evaluates models based on Long Short-Term Memory (LSTM) layers, SimpleRNN, Gated Recurrent Unit (GRU), and a Hybrid RNN, applied to the classification of five activities: Biking, Roller Skiing (R-Skiing), Running, Skiing, and Walking. The results show that sliding window data augmentation significantly enhances model performance, improving key metrics such as precision, recall, F1-score, and accuracy. Among the models tested, the Hybrid RNN and GRU models demonstrated the highest accuracy and generalization capabilities. Additionally, we tested several window and step sizes. The configuration with a larger window size (256) generally yielded better results. These findings are consistent with existing literature, highlighting the effectiveness of data augmentation and advanced RNN architectures in time series classification. The study highlights the importance of data augmentation in improving model robustness and provides valuable insights for future research and practical applications in various fields.

La modélisation par équations structurelles est très répandue dans des domaines très variés et nous l'appliquons pour la première fois en métrologie dans le traitement de données de comparaisons interlaboratoires. Les modèles à équations structurelles à variables latentes sont des modèles multivariés utilisés pour modéliser des relations de causalité entre des variables observées (les données). Le modèle s'applique dans le cas où les données peuvent être regroupées dans des blocs disjoints où chaque bloc définit un concept modélisé par une variable latente. La structure de corrélation des variables observées est ainsi résumée dans la structure de corrélation des variables latentes. Nous proposons une approche bayésienne des modèles à équations structurelles centrée sur l'analyse de la matrice de corrélation des variables latentes. Nous appliquons une expansion paramétrique à la matrice de corrélation des variables latentes afin de surmonter l'indétermination de l'échelle des variables latentes et d'améliorer la convergence de l'algorithme de Gibbs utilisé. La puissance de l'approche structurelle nous permet de proposer une modélisation riche et flexible des biais de mesure qui vient enrichir le calcul de la valeur de consensus et de son incertitude associée dans un cadre entièrement bayésien. Sous certaines hypothèses l'approche permet de manière innovante de calculer les contributions des variables de biais au biais des laboratoires. Plus généralement nous proposons un cadre bayésien pour l'amélioration de la qualité des mesures. Nous illustrons et montrons l'intérêt d'une modélisation structurelle des biais de mesure sur des comparaisons interlaboratoires en environnement. / Structural equation modelling is a widespread approach in a variety of domains and is first applied here to interlaboratory comparisons in metrology. Structural Equation Models with latent variables (SEM) are multivariate models used to model causality relationships in observed variables (the data). It is assumed that data can be grouped into separate blocks each describing a latent concept modelled by a latent variable. The correlation structure of the observed variables is transferred into the correlation structure of the latent variables. A Bayesian approach of SEM is proposed based on the analysis of the correlation matrix of latent variables using parameter expansion to overcome identifiability issues and improving the convergence of the Gibbs sampler. SEM is used as a powerful and flexible tool to model measurement bias with the aim of improving the reliability of the consensus value and its associated uncertainty in a fully Bayesian framework. The approach also allows to compute the contributions of the observed variables to the bias of the laboratories, under additional hypotheses. More generally a global Bayesian framework is proposed to improve the quality of measurements. The approach is illustrated on the structural equation modelling of measurement bias in interlaboratory comparisons in environment.

Hlavným zámerom tejto diplomovej práce bola automatická segmentácia lézií sklerózy multiplex na snímkoch MRI. V rámci práce boli otestované najnovšie metódy segmentácie s využitím hlbokých neurónových sietí a porovnané prístupy inicializácie váh sietí pomocou preneseného učenia (transfer learning) a samoriadeného učenia (self-supervised learning). Samotný problém automatickej segmentácie lézií sklerózy multiplex je veľmi náročný, a to primárne kvôli vysokej nevyváženosti datasetu (skeny mozgov zvyčajne obsahujú len malé množstvo poškodeného tkaniva). Ďalšou výzvou je manuálna anotácia týchto lézií, nakoľko dvaja rozdielni doktori môžu označiť iné časti mozgu ako poškodené a hodnota Dice Coefficient týchto anotácií je približne 0,86. Možnosť zjednodušenia procesu anotovania lézií automatizáciou by mohlo zlepšiť výpočet množstva lézií, čo by mohlo viesť k zlepšeniu diagnostiky individuálnych pacientov. Našim cieľom bolo navrhnutie dvoch techník využívajúcich transfer learning na predtrénovanie váh, ktoré by neskôr mohli zlepšiť výsledky terajších segmentačných modelov. Teoretická časť opisuje rozdelenie umelej inteligencie, strojového učenia a hlbokých neurónových sietí a ich využitie pri segmentácii obrazu. Následne je popísaná skleróza multiplex, jej typy, symptómy, diagnostika a liečba. Praktická časť začína predspracovaním dát. Najprv boli skeny mozgu upravené na rovnaké rozlíšenie s rovnakou veľkosťou voxelu. Dôvodom tejto úpravy bolo využitie troch odlišných datasetov, v ktorých boli skeny vytvárané rozličnými prístrojmi od rôznych výrobcov. Jeden dataset taktiež obsahoval lebku, a tak bolo nutné jej odstránenie pomocou nástroju FSL pre ponechanie samotného mozgu pacienta. Využívali sme 3D skeny (FLAIR, T1 a T2 modality), ktoré boli postupne rozdelené na individuálne 2D rezy a použité na vstup neurónovej siete s enkodér-dekodér architektúrou. Dataset na trénovanie obsahoval 6720 rezov s rozlíšením 192 x 192 pixelov (po odstránení rezov, ktorých maska neobsahovala žiadnu hodnotu). Využitá loss funkcia bola Combo loss (kombinácia Dice Loss s upravenou Cross-Entropy). Prvá metóda sa zameriavala na využitie predtrénovaných váh z ImageNet datasetu na enkodér U-Net architektúry so zamknutými váhami enkodéra, resp. bez zamknutia a následného porovnania s náhodnou inicializáciou váh. V tomto prípade sme použili len FLAIR modalitu. Transfer learning dokázalo zvýšiť sledovanú metriku z hodnoty približne 0,4 na 0,6. Rozdiel medzi zamknutými a nezamknutými váhami enkodéru sa pohyboval okolo 0,02. Druhá navrhnutá technika používala self-supervised kontext enkodér s Generative Adversarial Networks (GAN) na predtrénovanie váh. Táto sieť využívala všetky tri spomenuté modality aj s prázdnymi rezmi masiek (spolu 23040 obrázkov). Úlohou GAN siete bolo dotvoriť sken mozgu, ktorý bol prekrytý čiernou maskou v tvare šachovnice. Takto naučené váhy boli následne načítané do enkodéru na aplikáciu na náš segmentačný problém. Tento experiment nevykazoval lepšie výsledky, s hodnotou DSC 0,29 a 0,09 (nezamknuté a zamknuté váhy enkodéru). Prudké zníženie metriky mohlo byť spôsobené použitím predtrénovaných váh na vzdialených problémoch (segmentácia a self-supervised kontext enkodér), ako aj zložitosť úlohy kvôli nevyváženému datasetu.

This thesis addresses the problem where we want to apply machine learning over a small data set of multivariate time series. A challenge when classifying data is when the data set is small and overfitting is at risk. Augmentation of small data sets might avoid overfitting. The multivariate time series used in this project represent motion data of people with reconstructed ACLs and a control group. The approach was pairing motion data from the training set and using Euclidean Barycentric Averaging to create a new set of synthetic motion data so as to increase the size of the training set. The classifiers used were Dynamic Time Warping -One Nearest neighbour and Time Series Forest. In our example we found this way of increasing the training set a less productive strategy. We also found Time Series Forest to generally perform with higher accuracy on the chosen data sets, but there may be more effective augmentation strategies to avoid overfitting.

La modélisation par équations structurelles est très répandue dans des domaines très variés et nous l'appliquons pour la première fois en métrologie dans le traitement de données de comparaisons interlaboratoires. Les modèles à équations structurelles à variables latentes sont des modèles multivariés utilisés pour modéliser des relations de causalité entre des variables observées (les données). Le modèle s'applique dans le cas où les données peuvent être regroupées dans des blocs disjoints où chaque bloc définit un concept modélisé par une variable latente. La structure de corrélation des variables observées est ainsi résumée dans la structure de corrélation des variables latentes. Nous proposons une approche bayésienne des modèles à équations structurelles centrée sur l'analyse de la matrice de corrélation des variables latentes. Nous appliquons une expansion paramétrique à la matrice de corrélation des variables latentes afin de surmonter l'indétermination de l'échelle des variables latentes et d'améliorer la convergence de l'algorithme de Gibbs utilisé. La puissance de l'approche structurelle nous permet de proposer une modélisation riche et flexible des biais de mesure qui vient enrichir le calcul de la valeur de consensus et de son incertitude associée dans un cadre entièrement bayésien. Sous certaines hypothèses l'approche permet de manière innovante de calculer les contributions des variables de biais au biais des laboratoires. Plus généralement nous proposons un cadre bayésien pour l'amélioration de la qualité des mesures. Nous illustrons et montrons l'intérêt d'une modélisation structurelle des biais de mesure sur des comparaisons interlaboratoires en environnement. / Structural equation modelling is a widespread approach in a variety of domains and is first applied here to interlaboratory comparisons in metrology. Structural Equation Models with latent variables (SEM) are multivariate models used to model causality relationships in observed variables (the data). It is assumed that data can be grouped into separate blocks each describing a latent concept modelled by a latent variable. The correlation structure of the observed variables is transferred into the correlation structure of the latent variables. A Bayesian approach of SEM is proposed based on the analysis of the correlation matrix of latent variables using parameter expansion to overcome identifiability issues and improving the convergence of the Gibbs sampler. SEM is used as a powerful and flexible tool to model measurement bias with the aim of improving the reliability of the consensus value and its associated uncertainty in a fully Bayesian framework. The approach also allows to compute the contributions of the observed variables to the bias of the laboratories, under additional hypotheses. More generally a global Bayesian framework is proposed to improve the quality of measurements. The approach is illustrated on the structural equation modelling of measurement bias in interlaboratory comparisons in environment.

<p>A wide range of optoelectronic applications demand semiconductors optimized for purpose.</p> <p>My research focused on data-driven identification of ABX3 Halide perovskite compositions for optimum photovoltaic absorption in solar cells.</p> <p>I trained machine learning models on previously reported datasets of halide perovskite band gaps based on first principles computations performed at different fidelities.</p> <p>Using these, I identified mixtures of candidate constituents at the A, B or X sites of the perovskite supercell which leveraged how mixed perovskite band gaps deviate from the linear interpolations predicted by Vegard's law of mixing to obtain a selection of stable perovskites with band gaps in the ideal range of 1 to 2 eV for visible light spectrum absorption.</p> <p>These models predict the perovskite band gap using the composition and inherent elemental properties as descriptors.</p> <p>This enables accurate, high fidelity prediction and screening of the much larger chemical space from which the data samples were drawn.</p> <p><br></p> <p>I utilized a recently published density functional theory (DFT) dataset of more than 1300 perovskite band gaps from four different levels of theory, added to an experimental perovskite band gap dataset of \textasciitilde{}100 points, to train random forest regression (RFR), Gaussian process regression (GPR), and Sure Independence Screening and Sparsifying Operator (SISSO) regression models, with data fidelity added as one-hot encoded features.</p> <p>I found that RFR yields the best model with a band gap root mean square error of 0.12 eV on the total dataset and 0.15 eV on the experimental points.</p> <p>SISSO provided compound features and functions for direct prediction of band gap, but errors were larger than from RFR and GPR.</p> <p>Additional insights gained from Pearson correlation and Shapley additive explanation (SHAP) analysis of learned descriptors suggest the RFR models performed best because of (a) their focus on identifying and capturing relevant feature interactions and (b) their flexibility to represent nonlinear relationships between such interactions and the band gap.</p> <p>The best model was deployed for predicting experimental band gap of 37785 hypothetical compounds.</p> <p>Based on this, we identified 1251 stable compounds with band gap predicted to be between 1 and 2 eV at experimental accuracy, successfully narrowing the candidates to about 3% of the screened compositions.</p>

[pt] A partir de melhorias incrementais em uma conhecida rede neural convolucional (U-Net), diferentes técnicas são avaliadas quanto às suas performances na tarefa de segmentação semântica em imagens sísmicas. Mais especificamente, procura-se a identificação e delineamento de estruturas salinas no subsolo, o que é de grande relevância na indústria de óleo e gás para a exploração de petróleo em camadas pré-sal, por exemplo. Além disso, os desafios apresentados no tratamento destas imagens sísmicas se assemelham em muito aos encontrados em tarefas de áreas médicas como identificação de tumores e segmentação de tecidos, o que torna o estudo da tarefa em questão ainda mais valioso. Este trabalho pretende sugerir uma metodologia adequada de abordagem à tarefa e produzir redes neurais capazes de segmentar imagens sísmicas com bons resultados dentro das métricas utilizadas. Para alcançar estes objetivos, diferentes estruturas de redes, transferência de aprendizado e técnicas de aumentação de dados são testadas em dois datasets com diferentes níveis de complexidade. / [en] Through incremental improvements in a well-known convolutional neural network (U-Net), different techniques are evaluated regarding their performance on the task of semantic segmentation of seismic images. More specifically, the objective is the better identification and outline of subsurface salt structures, which is a task of great relevance for the oil and gas industry in the exploration of pre-salt layers, for example. Besides that application, the challenges imposed by the treatment of seismic images also resemble those found in medical fields like tumor detection and tissue segmentation, which makes the study of this task even more valuable. This work seeks to suggest a suitable methodology for the task and to yield neural networks that are capable of performing semantic segmentation of seismic images with good results regarding specific metrics. For that purpose, different network structures, transfer learning and data augmentation techniques are applied in two datasets with different levels of complexity.

No description available.

Perception, primarily through vision, is a vital human ability that informs decision-making and interactions with the world. Computer Vision, the field dedicated to emulating this human capability in computers, has witnessed transformative progress with the advent of artificial intelligence, particularly neural networks and deep learning. These technologies enable automatic feature learning, eliminating the need for laborious hand-crafted features. The increasing global demand for artificial intelligence applications across various industries, however, raises concerns about data privacy and access. This dissertation addresses these challenges by proposing solutions that leverage synthetic data to preserve privacy and enhance the robustness of computer vision algorithms. The primary objective of this dissertation is to reduce the dependence on real data for modern image processing algorithms by utilizing synthetic data generated through computer simulations. Synthetic data serves as a privacy-preserving alternative, enabling the generation of data in scenarios that are difficult or unsafe to replicate in the real world. While purely simulated data falls short of capturing the full complexity of reality, the dissertation explores methods to bridge the gap between synthetic and real data. The dissertation encompasses a comprehensive evaluation of the synthetic THEODORE dataset, focusing on object detection using Convolutional Neural Networks. Fine-tuning CNN architectures with synthetic data demonstrates remarkable performance improvements over relying solely on real-world data. Extending beyond person recognition, these architectures exhibit the ability to recognize various objects in real-world settings. This work also investigates real-time performance and the impact of barrel distortion in omnidirectional images, underlining the potential of using synthetic data. Furthermore, the dissertation introduces two unsupervised domain adaptation methods tailored for anchorless object detection within the CenterNet architecture. The methods effectively reduce the domain gap when synthetic omnidirectional images serve as the source domain, and real images act as the target domain. Qualitative assessments highlight the advantages of these methods in reducing noise and enhancing detection accuracy. The dissertation concludes with creating an application within the Ambient Assisted Living context to realize the concepts. This application encompasses indoor localization heatmaps, human pose estimation, and activity recognition. The methodology leverages synthetically generated data, unique object identifiers, and rotated bounding boxes to enhance tracking in omnidirectional images. Importantly, the system is designed to operate without compromising privacy or using sensitive images, aligning with the growing concerns of data privacy and access in artificial intelligence applications. / Die Wahrnehmung, insbesondere durch das Sehen, ist eine entscheidende menschliche Fähigkeit, die die Entscheidungsfindung und die Interaktion mit der Welt beeinflusst. Die Computer Vision, das Fachgebiet, das sich der Nachahmung dieser menschlichen Fähigkeit in Computern widmet, hat mit dem Aufkommen künstlicher Intelligenz, insbesondere neuronaler Netzwerke und tiefem Lernen, eine transformative Entwicklung erlebt. Diese Technologien ermöglichen das automatische Erlernen von Merkmalen und beseitigen die Notwendigkeit mühsamer, handgefertigter Merkmale. Die steigende weltweite Nachfrage nach Anwendungen künstlicher Intelligenz in verschiedenen Branchen wirft jedoch Bedenken hinsichtlich des Datenschutzes und des Datenzugriffs auf. Diese Dissertation begegnet diesen Herausforderungen, indem sie Lösungen vorschlägt, die auf synthetischen Daten basieren, um die Privatsphäre zu wahren und die Robustheit von Computer-Vision Algorithmen zu steigern. Das Hauptziel dieser Dissertation besteht darin, die Abhängigkeit von realen Daten für moderne Bildverarbeitungsalgorithmen durch die Verwendung von synthetischen Daten zu reduzieren, die durch Computersimulationen generiert werden. Synthetische Daten dienen als datenschutzfreundliche Alternative und ermöglichen die Generierung von Daten in Szenarien, die schwer oder unsicher in der realen Welt nachzustellen sind. Obwohl rein simulierte Daten die volle Komplexität der Realität nicht erfassen, erforscht die Dissertation Methoden zur Überbrückung der Kluft zwischen synthetischen und realen Daten. Die Dissertation umfasst eine Evaluation des synthetischen THEODORE-Datensatzes mit dem Schwerpunkt auf der Objekterkennung mithilfe von Convolutional Neural Networks. Das Feinabstimmen dieser Architekturen mit synthetischen Daten zeigt bemerkenswerte Leistungssteigerungen im Vergleich zur ausschließlichen Verwendung von realen Daten. Über die Erkennung von Personen hinaus zeigen diese Architekturen die Fähigkeit, verschiedene Objekte in realen Umgebungen zu erkennen. Untersucht wird auch die Echtzeit-Performance und der Einfluss der tonnenförmigen Verzerrung in omnidirektionalen Bildern und betont das Potenzial der Verwendung synthetischer Daten. Darüber hinaus führt die Dissertation zwei nicht überwachte Domänenanpassungsmethoden ein, die speziell für die ankerlose Objekterkennung in der CenterNetArchitektur entwickelt wurden. Die Methoden reduzieren effektiv die Domänenlücke, wenn synthetische omnidirektionale Bilder als Quelldomäne und reale Bilder als Zieldomäne dienen. Qualitative Bewertungen heben die Vorteile dieser Methoden bei der Reduzierung von Störungen und der Verbesserung der Erkennungsgenauigkeit hervor. Die Dissertation schließt mit der Entwicklung einer Anwendung im Kontext von Ambient Assisted Living zur Umsetzung der Konzepte. Diese Anwendung umfasst Innenlokalisierungskarten, die Schätzung der menschlichen Körperhaltung und die Erkennung von Aktivitäten. Die Methodologie nutzt synthetisch generierte Daten, eindeutige Objektidentifikatoren und rotierte Begrenzungsrahmen, um die Verfolgung in omnidirektionalen Bildern zu verbessern. Wichtig ist, dass das System entwickelt wurde, um ohne Beeinträchtigung der Privatsphäre oder Verwendung sensibler Bilder zu arbeiten, was den wachsenden Bedenken hinsichtlich des Datenschutzes und des Zugriffs auf Daten in Anwendungen künstlicher Intelligenz entspricht.