Semantic segmentation is the process of assigning a specific class label to each pixel in an image. There are multiple areas of use for semantic segmentation of remote sensing images, including climate change studies and urban planning and development. When training a network to perform semantic segmentation in a supervised manner, annotated data is crucial, and annotating satellite images is an expensive and time-consuming task. A resolution to this issue might be self-supervised learning. Training a pretext task on a large unlabeled dataset, and a downstream task on a smaller labeled dataset, could mitigate the need for large amounts of labeled data. In this thesis, the use of self-supervised learning for semantic segmentation of remote sensing data is investigated and compared to the traditional use of supervised pre-training using ImageNet. Two different methods of self-supervised learning are evaluated, a reconstructive method and a contrastive method. Furthermore, whether including modalities unique to remote sensing data yields greater performance for semantic segmentation is investigated. The findings indicate that self-supervised learning with in-domain data shows significant potential. While the performance of models pre-trained using self-supervised learning on remote sensing data, does not surpass that of pre-trained models using supervised learning on ImageNet, it achieves a comparable level. This is notable given the substantially smaller training data used. However, in cases where the in-domain dataset is small — as in this thesis with approximately 20,000 images — leveraging ImageNet for pre-training is preferable. Furthermore, self-supervised learning demonstrates promise as a more effective pre-training approach compared to supervised learning, when both methods are trained on ImageNet. The reconstructive method proves more suitable for semantic segmentation of remote sensing data compared to the contrastive method, and incorporating modalities unique to remote sensing further enhances performance.

Cette thèse tente de construire de meilleurs agents d'apprentissage par renforcement (RL) en tirant parti de l'apprentissage auto-supervisé. Il se présente sous la forme d'une thèse par article qui contient trois travaux. Dans le premier article, nous construisons un benchmark basé sur les jeux Atari pour évaluer systématiquement les méthodes d'apprentissage auto-supervisé dans les environnements RL. Nous comparons un éventail de ces méthodes à travers une suite de tâches de sondage pour identifier leurs forces et leurs faiblesses. Nous montrons en outre qu'une nouvelle méthode contrastive ST-DIM excelle à capturer la plupart des facteurs génératifs dans les environnements étudiés, sans avoir besoin de s'appuyer sur des étiquettes ou des récompenses. Dans le deuxième article, nous proposons des représentations auto-prédictives (SPR) qui apprennent un modèle latent auto-supervisé de la dynamique de l'environnement parallèlement à la résolution de la tâche RL en cours. Nous montrons que SPR réalise des améliorations spectaculaires dans l'état de l'art sur le benchmark Atari 100k difficile où les agents n'ont droit qu'à 2 heures d'expérience en temps réel. Le troisième article étudie le rôle de la RL basée sur un modèle et de l'apprentissage auto-supervisé dans le contexte de la généralisation en RL. Grâce à des contrôles minutieux, nous montrons que la planification et l'apprentissage de représentation basé sur un modèle contribuent tous deux à une meilleure généralisation pour l'agent Muzero. Nous améliorons encore MuZero avec des objectifs d'apprentissage auto-supervisés auxiliaires, et montrons que cet agent MuZero++ obtient des résultats de pointe sur les benchmarks Procgen et Metaworld. / This thesis tries to build better Reinforcement Learning (RL) agents by leveraging self-supervised learning. It is presented as a thesis by article that contains three pieces of work. In the first article, we construct a benchmark based on Atari games to systematically evaluate self-supervised learning methods in RL environments. We compare an array of such methods across a suite of probing tasks to identify their strengths and weaknesses. We further show that a novel contrastive method ST-DIM excels at capturing most generative factors in the studied environments, without needing to rely on labels or rewards. In the second article, we propose Self-Predictive Representations (SPR) that learns a self-supervised latent model of the environment dynamics alongside solving the RL task at hand. We show that SPR achieves dramatic improvements in state-of-the-art on the challenging Atari 100k benchmark where agents are allowed only 2 hours of real-time experience. The third article studies the role of model-based RL and self-supervised learning in the context of generalization in RL. Through careful controls, we show that planning and model-based representation learning both contribute towards better generalization for the Muzero agent. We further improve MuZero with auxiliary self-supervised learning objectives, and show that this MuZero++ agent achieves state-of-the-art results on the Procgen and Metaworld benchmarks.

Établir des méthodes d'évaluation pour les systèmes d'intelligence artificielle (IA) est une étape importante pour précisément connaître leurs limites et ainsi prévenir les dommages qu'ils pourraient causer et savoir quels aspects devraient être améliorés. Cela nécessite d'être en mesure de dresser des portraits précis des limitations associées à un système d'IA donné. Cela demande l'accès à des outils et des principes fiables, transparent, à jour et faciles à utiliser. Malheureusement, la plupart des méthodes d'évaluation utilisées à ce jour ont un retard significatif par rapport aux performances toujours croissantes des réseaux de neurones artificiels. Dans cette thèse par articles, je présente des méthodes et des principes d'évaluation plus rigoureux pour obtenir une meilleur compréhension des réseaux de neurones et de leurs limitations. Dans le premier article, je présente Representation Conditional Diffusion Model (RCDM), une méthode d'évaluation à l'état de l'art qui permet, à partir d'une représentation donnée -- par exemple les activations d'une couche donnée d'un réseau de neurones artificiels -- de générer une image. En utilisant les dernières avancées dans la génération d'images, RCDM permet aux chercheur·euse·s de visualiser l'information contenue à l'intérieur d'une représentation. Dans le deuxième article, j'introduis la régularisation par Guillotine qui est une technique bien connue dans la littérature sur l'apprentissage par transfert mais qui se présente différemment dans la littérature sur l'auto-apprentissage. Pour améliorer la généralisation à travers différentes tâches, on montre qu'il est important d'évaluer un modèle en coupant un certain nombre de couches. Dans le troisième article, j'introduis le score DéjaVu qui quantifie à quel point un réseau de neurones a mémorisé les données d'entraînement. Ce score utilise une petite partie d'une image d'entraînement puis évalue quelles informations il est possible d'inférer à propos du reste de l'image. Dans le dernier article, je présente les jeux de données photo-réalistes PUG (Photorealistic Unreal Graphics) que nous avons développés. Au contraire de données réelles, pour lesquelles générer des annotations est un processus coûteux, l'utilisation de données synthétiques offre un contrôle total sur la scène générée et sur les annotations. On utilise un moteur de jeux vidéo qui permet la synthèse d'images photo-réalistes de haute qualité, afin d'évaluer la robustesse d'un réseau de neurones pré-entraîné, ceci sans avoir besoin d'adapter ce réseau avec un entraînement additionnel. / Carefully designing benchmarks to evaluate the safety of Artificial Intelligent (AI) agents is a much-needed step to precisely know the limits of their capabilities and thus prevent potential damages they could cause if used beyond these limits. Researchers and engineers should be able to draw precise pictures of the failure modes of a given AI system and find ways to mitigate them. Drawing such portraits requires reliable tools and principles that are transparent, up-to-date, and easy to use by practitioners. Unfortunately, most of the benchmark tools used in research are often outdated and quickly fall behind the fast pace of improvement of the capabilities of deep neural networks. In this thesis by article, I focus on establishing more fine-grained evaluation methods and principles to gain a better understanding of deep neural networks and their limitations. In the first article, I present Representation Conditional Diffusion Model (RCDM), a state-of-the-art visualization method that can map any deep neural network representation to the image space. Using the latest advances in generative modeling, RCDM sheds light on what is learned by deep neural networks by allowing practitioners to visualize the richness of a given representation. In the second article, I (re)introduce Guillotine Regularization (GR) -- a trick that has been used for a long time in transfer learning -- from a novel understanding and viewpoint grounded in the self-supervised learning outlook. We show that evaluating a model by removing its last layers is important to ensure better generalization across different downstream tasks. In the third article, I introduce the DejaVu score which quantifies how much models are memorizing their training data. This score relies on leveraging partial information from a given image such as a crop, and evaluates how much information one can retrieve about the entire image based on only this partial content. In the last article, I introduce the Photorealistic Unreal Graphics (PUG) datasets and benchmarks. In contrast to real data for which getting annotations is often a costly and long process, synthetic data offers complete control of the elements in the scene and labeling. In this work, we leverage a powerful game engine that produces high-quality and photorealistic images to evaluate the robustness of pre-trained neural networks without additional finetuning.

Microtomography, or microCT, is an x-ray imaging modality that provides volumetric data of an object's internal structure with microscale resolution, making it suitable for scanning small, highly detailed objects. The microCT image quality is limited by quantum noise, which can be reduced by increasing the scan time. This complicates the scanning both of dynamic processes and, due to the increased radiation dose, dose-sensitive samples. A recently proposed method for improved dose- or time-limited scanning is Noise2Inverse, a framework for denoising data in tomography and linear inverse problems by training a self-supervised convolutional neural network. This work implements Noise2Inverse for denoising lab-based cone-beam microCT data and compares it to both supervised neural networks and more traditional filtering methods. While some trade-off in spatial resolution is observed, the method outperforms traditional filtering methods and matches supervised denoising in quantitative and qualitative evaluations of image quality. Additionally, a segmentation task is performed to show that denoising the data can aid in practical tasks. / Mikrotomografi, eller mikroCT, är en röntgenmetod som avbildar små objekt i tre dimensioner med upplösning på mikrometernivå, vilket möjligör avbildning av små och högdetaljerade objekt. Bildkvaliteten vid mikroCT begränsas av kvantbrus, vilket kan minskas genom att öka skanningstiden. Detta försvårar avbildning av dynamiska processer och, på grund av den ökade stråldosen, doskänsliga objekt. En metod som tros kunna förbättra dos- eller tidsbegränsad avbildning är Noise2Inverse, ett ramverk för brusreducering av tomografisk data genom träning av ett självövervakat faltningsnätverk, och jämförs med både övervakade neuronnät och mer traditionella filtermetoder. Noise2Inverse implementaras i detta arbete för brusreducering av data från ett labb-baserat mikroCT-system med cone beam-geometri. En viss reducering i spatiell upplösning observeras, men metoden överträffar traditionella filtermetoder och matchar övervakade neuronnät i kvantitativa och kvalitativa utvärderingar av bildkvalitet. Dessutom visas att metoden går att använda för att förbätta resultat från bildsegmentering.

Extracting useful knowledge from data sets is a key concept in modern information systems. Consequently, the need of efficient techniques to extract the desired knowledge has been growing over time. Machine learning is a research field dedicated to the development of techniques capable of enabling a machine to \"learn\" from data. Many techniques have been proposed so far, but there are still issues to be unveiled specially in interdisciplinary research. In this thesis, we explore the advantages of network data representation to develop machine learning techniques based on dynamical processes on networks. The network representation unifies the structure, dynamics and functions of the system it represents, and thus is capable of capturing the spatial, topological and functional relations of the data sets under analysis. We develop network-based techniques for the three machine learning paradigms: supervised, semi-supervised and unsupervised. The random walk dynamical process is used to characterize the access of unlabeled data to data classes, configuring a new heuristic we call ease of access in the supervised paradigm. We also propose a classification technique which combines the high-level view of the data, via network topological characterization, and the low-level relations, via similarity measures, in a general framework. Still in the supervised setting, the modularity and Katz centrality network measures are applied to classify multiple observation sets, and an evolving network construction method is applied to the dimensionality reduction problem. The semi-supervised paradigm is covered by extending the ease of access heuristic to the cases in which just a few labeled data samples and many unlabeled samples are available. A semi-supervised technique based on interacting forces is also proposed, for which we provide parameter heuristics and stability analysis via a Lyapunov function. Finally, an unsupervised network-based technique uses the concepts of pinning control and consensus time from dynamical processes to derive a similarity measure used to cluster data. The data is represented by a connected and sparse network in which nodes are dynamical elements. Simulations on benchmark data sets and comparisons to well-known machine learning techniques are provided for all proposed techniques. Advantages of network data representation and dynamical processes for machine learning are highlighted in all cases / A extração de conhecimento útil a partir de conjuntos de dados é um conceito chave em sistemas de informação modernos. Por conseguinte, a necessidade de técnicas eficientes para extrair o conhecimento desejado vem crescendo ao longo do tempo. Aprendizado de máquina é uma área de pesquisa dedicada ao desenvolvimento de técnicas capazes de permitir que uma máquina \"aprenda\" a partir de conjuntos de dados. Muitas técnicas já foram propostas, mas ainda há questões a serem reveladas especialmente em pesquisas interdisciplinares. Nesta tese, exploramos as vantagens da representação de dados em rede para desenvolver técnicas de aprendizado de máquina baseadas em processos dinâmicos em redes. A representação em rede unifica a estrutura, a dinâmica e as funções do sistema representado e, portanto, é capaz de capturar as relações espaciais, topológicas e funcionais dos conjuntos de dados sob análise. Desenvolvemos técnicas baseadas em rede para os três paradigmas de aprendizado de máquina: supervisionado, semissupervisionado e não supervisionado. O processo dinâmico de passeio aleatório é utilizado para caracterizar o acesso de dados não rotulados às classes de dados configurando uma nova heurística no paradigma supervisionado, a qual chamamos de facilidade de acesso. Também propomos uma técnica de classificação de dados que combina a visão de alto nível dos dados, por meio da caracterização topológica de rede, com relações de baixo nível, por meio de medidas de similaridade, em uma estrutura geral. Ainda no aprendizado supervisionado, as medidas de rede modularidade e centralidade Katz são aplicadas para classificar conjuntos de múltiplas observações, e um método de construção evolutiva de rede é aplicado ao problema de redução de dimensionalidade. O paradigma semissupervisionado é abordado por meio da extensão da heurística de facilidade de acesso para os casos em que apenas algumas amostras de dados rotuladas e muitas amostras não rotuladas estão disponíveis. É também proposta uma técnica semissupervisionada baseada em forças de interação, para a qual fornecemos heurísticas para selecionar parâmetros e uma análise de estabilidade mediante uma função de Lyapunov. Finalmente, uma técnica não supervisionada baseada em rede utiliza os conceitos de controle pontual e tempo de consenso de processos dinâmicos para derivar uma medida de similaridade usada para agrupar dados. Os dados são representados por uma rede conectada e esparsa na qual os vértices são elementos dinâmicos. Simulações com dados de referência e comparações com técnicas de aprendizado de máquina conhecidas são fornecidos para todas as técnicas propostas. As vantagens da representação de dados em rede e de processos dinâmicos para o aprendizado de máquina são evidenciadas em todos os casos

Extracting useful knowledge from data sets is a key concept in modern information systems. Consequently, the need of efficient techniques to extract the desired knowledge has been growing over time. Machine learning is a research field dedicated to the development of techniques capable of enabling a machine to \"learn\" from data. Many techniques have been proposed so far, but there are still issues to be unveiled specially in interdisciplinary research. In this thesis, we explore the advantages of network data representation to develop machine learning techniques based on dynamical processes on networks. The network representation unifies the structure, dynamics and functions of the system it represents, and thus is capable of capturing the spatial, topological and functional relations of the data sets under analysis. We develop network-based techniques for the three machine learning paradigms: supervised, semi-supervised and unsupervised. The random walk dynamical process is used to characterize the access of unlabeled data to data classes, configuring a new heuristic we call ease of access in the supervised paradigm. We also propose a classification technique which combines the high-level view of the data, via network topological characterization, and the low-level relations, via similarity measures, in a general framework. Still in the supervised setting, the modularity and Katz centrality network measures are applied to classify multiple observation sets, and an evolving network construction method is applied to the dimensionality reduction problem. The semi-supervised paradigm is covered by extending the ease of access heuristic to the cases in which just a few labeled data samples and many unlabeled samples are available. A semi-supervised technique based on interacting forces is also proposed, for which we provide parameter heuristics and stability analysis via a Lyapunov function. Finally, an unsupervised network-based technique uses the concepts of pinning control and consensus time from dynamical processes to derive a similarity measure used to cluster data. The data is represented by a connected and sparse network in which nodes are dynamical elements. Simulations on benchmark data sets and comparisons to well-known machine learning techniques are provided for all proposed techniques. Advantages of network data representation and dynamical processes for machine learning are highlighted in all cases / A extração de conhecimento útil a partir de conjuntos de dados é um conceito chave em sistemas de informação modernos. Por conseguinte, a necessidade de técnicas eficientes para extrair o conhecimento desejado vem crescendo ao longo do tempo. Aprendizado de máquina é uma área de pesquisa dedicada ao desenvolvimento de técnicas capazes de permitir que uma máquina \"aprenda\" a partir de conjuntos de dados. Muitas técnicas já foram propostas, mas ainda há questões a serem reveladas especialmente em pesquisas interdisciplinares. Nesta tese, exploramos as vantagens da representação de dados em rede para desenvolver técnicas de aprendizado de máquina baseadas em processos dinâmicos em redes. A representação em rede unifica a estrutura, a dinâmica e as funções do sistema representado e, portanto, é capaz de capturar as relações espaciais, topológicas e funcionais dos conjuntos de dados sob análise. Desenvolvemos técnicas baseadas em rede para os três paradigmas de aprendizado de máquina: supervisionado, semissupervisionado e não supervisionado. O processo dinâmico de passeio aleatório é utilizado para caracterizar o acesso de dados não rotulados às classes de dados configurando uma nova heurística no paradigma supervisionado, a qual chamamos de facilidade de acesso. Também propomos uma técnica de classificação de dados que combina a visão de alto nível dos dados, por meio da caracterização topológica de rede, com relações de baixo nível, por meio de medidas de similaridade, em uma estrutura geral. Ainda no aprendizado supervisionado, as medidas de rede modularidade e centralidade Katz são aplicadas para classificar conjuntos de múltiplas observações, e um método de construção evolutiva de rede é aplicado ao problema de redução de dimensionalidade. O paradigma semissupervisionado é abordado por meio da extensão da heurística de facilidade de acesso para os casos em que apenas algumas amostras de dados rotuladas e muitas amostras não rotuladas estão disponíveis. É também proposta uma técnica semissupervisionada baseada em forças de interação, para a qual fornecemos heurísticas para selecionar parâmetros e uma análise de estabilidade mediante uma função de Lyapunov. Finalmente, uma técnica não supervisionada baseada em rede utiliza os conceitos de controle pontual e tempo de consenso de processos dinâmicos para derivar uma medida de similaridade usada para agrupar dados. Os dados são representados por uma rede conectada e esparsa na qual os vértices são elementos dinâmicos. Simulações com dados de referência e comparações com técnicas de aprendizado de máquina conhecidas são fornecidos para todas as técnicas propostas. As vantagens da representação de dados em rede e de processos dinâmicos para o aprendizado de máquina são evidenciadas em todos os casos

No description available.

No description available.

No description available.

[pt] Prever com precisão a resposta a drogas para uma dada amostra baseado em características moleculares pode ajudar a otimizar o desenvolvimento de drogas e explicar mecanismos por trás das respostas aos tratamentos. Nessa dissertação, dois estudos de caso foram gerados, cada um aplicando diferentes dados genômicos para a previsão de resposta a drogas. O estudo de caso 1 avaliou dados de perfis de metilação de DNA como um tipo de característica molecular que se sabe ser responsável por causar tumorigênese e modular a resposta a tratamentos. Usando perfis de metilação de 987 linhagens celulares do genoma completo na base de dados Genomics of Drug Sensitivity in Cancer (GDSC), utilizamos algoritmos de aprendizado de máquina para avaliar o potencial preditivo de respostas citotóxicas para oito drogas contra o câncer. Nós comparamos a performance de cinco algoritmos de classificação e quatro algoritmos de regressão representando metodologias diversas, incluindo abordagens tree-, probability-, kernel-, ensemble- e distance-based. Aplicando sub-amostragem artificial em graus variados, essa pesquisa procura avaliar se o treinamento baseado em resultados relativamente extremos geraria melhoria no desempenho. Ao utilizar algoritmos de classificação e de regressão para prever respostas discretas ou contínuas, respectivamente, nós observamos consistentemente excelente desempenho na predição quando os conjuntos de treinamento e teste consistiam em dados de linhagens celulares. Algoritmos de classificação apresentaram melhor desempenho quando nós treinamos os modelos utilizando linhagens celulares com valores de resposta a drogas relativamente extremos, obtendo valores de area-under-the-receiver-operating-characteristic-curve de até 0,97. Os algoritmos de regressão tiveram melhor desempenho quando treinamos os modelos utilizado o intervalo completo de valores de resposta às drogas, apesar da dependência das métricas de desempenho utilizadas. O estudo de caso 2 avaliou dados de RNA-seq, dados estes comumente utilizados no estudo da eficácia de drogas. Aplicando uma abordagem de aprendizado semi-supervisionado, essa pesquisa busca avaliar o impacto da combinação de dados rotulados e não-rotulados para melhorar a predição do modelo. Usando dados rotulados de RNA-seq do genoma completo de uma média de 125 amostras de tumor AML rotuladas da base de dados Beat AML (separados por tipos de droga) e 151 amostras de tumor AML não-rotuladas na base de dados The Cancer Genome Atlas (TCGA), utilizamos uma estrutura de modelo semi-supervisionado para prever respostas citotóxicas para quatro drogas contra câncer. Modelos semi-supervisionados foram gerados, avaliando várias combinações de parâmetros e foram comparados com os algoritmos supervisionados de classificação. / [en] Accurately predicting drug responses for a given sample based on molecular features may help to optimize drug-development pipelines and explain mechanisms behind treatment responses. In this dissertation, two case studies were generated, each applying different genomic data to predict drug response. Case study 1 evaluated DNA methylation profile data as one type of molecular feature that is known to drive tumorigenesis and modulate treatment responses. Using genome-wide, DNA methylation profiles from 987 cell lines in the Genomics of Drug Sensitivity in Cancer (GDSC) database, we used machine-learning algorithms to evaluate the potential to predict cytotoxic responses for eight anti-cancer drugs. We compared the performance of five classification algorithms and four regression algorithms representing diverse methodologies, including tree-, probability-, kernel-, ensemble- and distance-based approaches. By applying artificial subsampling in varying degrees, this research aims to understand whether training based on relatively extreme outcomes would yield improved performance. When using classification or regression algorithms to predict discrete or continuous responses, respectively, we consistently observed excellent predictive performance when the training and test sets consisted of cell-line data. Classification algorithms performed best when we trained the models using cell lines with relatively extreme drug-response values, attaining area-under-the-receiver-operating-characteristic-curve values as high as 0.97. The regression algorithms performed best when we trained the models using the full range of drug-response values, although this depended on the performance metrics we used. Case study 2 evaluated RNA-seq data as one of the most popular molecular data used to study drug efficacy. By applying a semi-supervised learning approach, this research aimed to understand the impact of combining labeled and unlabeled data to improve model prediction. Using genome-wide RNA-seq labeled data from an average of 125 AML tumor samples in the Beat AML database (varying by drug type) and 151 unlabeled AML tumor samples in The Cancer Genome Atlas (TCGA) database, we used a semi-supervised model structure to predict cytotoxic responses for four anti-cancer drugs. Semi-supervised models were generated, while assessing several parameter combinations and were compared against supervised classification algorithms.