Multivariate Gaussians: We provide sample complexity upper bounds for semi-agnostically learning multivariate Gaussians under the constraint of approximate differential privacy. These are the first  finite sample upper bounds for general Gaussians which do not impose restrictions on the parameters of the distribution. Our bounds are near-optimal in the case when the covariance is known to be the identity, and conjectured to be near-optimal in the general case. From a technical standpoint, we provide analytic tools for arguing the existence of global "locally small" covers from local covers of the space. These are exploited using modifications of recent techniques for for differentially private hypothesis selection. Mixtures of Gaussians: We consider the problem of learning mixtures of Gaussians under the constraint of approximate differential privacy. We provide the first sample complexity upper bounds for privately learning mixtures of unbounded axis-aligned (or even unbounded univariate) Gaussians. To prove our results, we design a new technique for privately learning mixture distributions. A class of distributions F is said to be list-decodable if there is an algorithm that, given "heavily corrupted" samples from a distribution f in F, outputs a list of distributions, H, such that one of the distributions in H approximates f. We show that if F is privately list-decodable then we can privately learn mixtures of distributions in F. Finally, we show axis-aligned Gaussian distributions are privately list-decodable, thereby proving mixtures of such distributions are privately learnable. / Thesis / Master of Science (MSc) / Is it possible to estimate an unknown probability distribution given random samples from it? This is a fundamental problem known as distribution learning (or density estimation) that has been studied by statisticians for decades, and in recent years has become a topic of interest for computer scientists. While distribution learning is a mature and well understood problem, in many cases the samples (or data) we observe may consist of sensitive information belonging to individuals and well-known solutions may inadvertently result in the leakage of private information. In this thesis we study distribution learning under the assumption that the data is generated from high-dimensional Gaussians (or their mixtures) with the aim of understanding how many samples an algorithm needs before it can guarantee a good estimate. Furthermore, to protect against leakage of private information, we consider approaches that satisfy differential privacy — the gold standard for modern private data analysis.

For many types of learners one can compute the statistically 'optimal' way to select data. We review how these techniques have been used with feedforward neural networks. We then show how the same principles may be used to select data for two alternative, statistically-based learning architectures: mixtures of Gaussians and locally weighted regression. While the techniques for neural networks are expensive and approximate, the techniques for mixtures of Gaussians and locally weighted regression are both efficient and accurate.

The learning process consists of different steps: building a Training Set (TS), training the system, testing its behaviour and finally classifying unknown objects. When using a distance based rule as a classifier, i.e. 1-Nearest Neighbour (1-NN), the first step (building a training set) includes editing and condensing data. The main reason for that is that the rules based on distance need many time to classify each unlabelled sample, x, as each distance from x to each point in the training set should be calculated. So, the more reduced the training set, the shorter the time needed for each new classification process. This thesis is mainly focused on building a training set from some already given data, and specially on condensing it; however different classification techniques are also compared.The aim of any condensing technique is to obtain a reduced training set in order to spend as few time as possible in classification. All that without a significant loss in classification accuracy. Somenew approaches to training set size reduction based on prototypes are presented. These schemes basically consist of defining a small number of prototypes that represent all the original instances. That includes those approaches that select among the already existing examples (selective condensing algorithms), and those which generate new representatives (adaptive condensing algorithms).Those new reduction techniques are experimentally compared to some traditional ones, for data represented in feature spaces. In order to test them, the classical 1-NN rule is here applied. However, other classifiers (fast classifiers) have been considered here, as linear and quadratic ones constructed in dissimilarity spaces based on prototypes, in order to realize how editing and condensing concepts work for this different family of classifiers.Although the goal of the algorithms proposed in this thesis is to obtain a strongly reduced set of representatives, the performance is empirically evaluated over eleven real data sets by comparing not only the reduction rate but also the classification accuracy with those of other condensing techniques. Therefore, the ultimate aim is not only to find a strongly reduced set, but also a balanced one.Several ways to solve the same problem could be found. So, in the case of using a rule based on distance as a classifier, not only the option of reducing the training set can be afford. A different family of approaches consists of applying several searching methods. Therefore, results obtained by the use of the algorithms here presented are compared in terms of classification accuracy and time, to several efficient search techniques.Finally, the main contributions of this PhD report could be briefly summarised in four principal points. Firstly, two selective algorithms based on the idea of surrounding neighbourhood. They obtain better results than other algorithms presented here, as well as better than other traditional schemes. Secondly, a generative approach based on mixtures of Gaussians. It presents better results in classification accuracy and size reduction than traditional adaptive algorithms, and similar to those of the LVQ. Thirdly, it is shown that classification rules other than the 1-NN can be used, even leading to better results. And finally, it is deduced from the experiments carried on, that with some databases (as the ones used here) the approaches here presented execute the classification processes in less time that the efficient search techniques.

Malgré des progrès constants en termes de capacité de calcul, mémoire et quantité de données disponibles, les algorithmes d'apprentissage machine doivent se montrer efficaces dans l'utilisation de ces ressources. La minimisation des coûts est évidemment un facteur important, mais une autre motivation est la recherche de mécanismes d'apprentissage capables de reproduire le comportement d'êtres intelligents. Cette thèse aborde le problème de l'efficacité à travers plusieurs articles traitant d'algorithmes d'apprentissage variés : ce problème est vu non seulement du point de vue de l'efficacité computationnelle (temps de calcul et mémoire utilisés), mais aussi de celui de l'efficacité statistique (nombre d'exemples requis pour accomplir une tâche donnée). Une première contribution apportée par cette thèse est la mise en lumière d'inefficacités statistiques dans des algorithmes existants. Nous montrons ainsi que les arbres de décision généralisent mal pour certains types de tâches (chapitre 3), de même que les algorithmes classiques d'apprentissage semi-supervisé à base de graphe (chapitre 5), chacun étant affecté par une forme particulière de la malédiction de la dimensionalité. Pour une certaine classe de réseaux de neurones, appelés réseaux sommes-produits, nous montrons qu'il peut être exponentiellement moins efficace de représenter certaines fonctions par des réseaux à une seule couche cachée, comparé à des réseaux profonds (chapitre 4). Nos analyses permettent de mieux comprendre certains problèmes intrinsèques liés à ces algorithmes, et d'orienter la recherche dans des directions qui pourraient permettre de les résoudre. Nous identifions également des inefficacités computationnelles dans les algorithmes d'apprentissage semi-supervisé à base de graphe (chapitre 5), et dans l'apprentissage de mélanges de Gaussiennes en présence de valeurs manquantes (chapitre 6). Dans les deux cas, nous proposons de nouveaux algorithmes capables de traiter des ensembles de données significativement plus grands. Les deux derniers chapitres traitent de l'efficacité computationnelle sous un angle différent. Dans le chapitre 7, nous analysons de manière théorique un algorithme existant pour l'apprentissage efficace dans les machines de Boltzmann restreintes (la divergence contrastive), afin de mieux comprendre les raisons qui expliquent le succès de cet algorithme. Finalement, dans le chapitre 8 nous présentons une application de l'apprentissage machine dans le domaine des jeux vidéo, pour laquelle le problème de l'efficacité computationnelle est relié à des considérations d'ingénierie logicielle et matérielle, souvent ignorées en recherche mais ô combien importantes en pratique. / Despite constant progress in terms of available computational power, memory and amount of data, machine learning algorithms need to be efficient in how they use them. Although minimizing cost is an obvious major concern, another motivation is to attempt to design algorithms that can learn as efficiently as intelligent species. This thesis tackles the problem of efficient learning through various papers dealing with a wide range of machine learning algorithms: this topic is seen both from the point of view of computational efficiency (processing power and memory required by the algorithms) and of statistical efficiency (n umber of samples necessary to solve a given learning task).The first contribution of this thesis is in shedding light on various statistical inefficiencies in existing algorithms. Indeed, we show that decision trees do not generalize well on tasks with some particular properties (chapter 3), and that a similar flaw affects typical graph-based semi-supervised learning algorithms (chapter 5). This flaw is a form of curse of dimensionality that is specific to each of these algorithms. For a subclass of neural networks, called sum-product networks, we prove that using networks with a single hidden layer can be exponentially less efficient than when using deep networks (chapter 4). Our analyses help better understand some inherent flaws found in these algorithms, and steer research towards approaches that may potentially overcome them. We also exhibit computational inefficiencies in popular graph-based semi-supervised learning algorithms (chapter 5) as well as in the learning of mixtures of Gaussians with missing data (chapter 6). In both cases we propose new algorithms that make it possible to scale to much larger datasets. The last two chapters also deal with computational efficiency, but in different ways. Chapter 7 presents a new view on the contrastive divergence algorithm (which has been used for efficient training of restricted Boltzmann machines). It provides additional insight on the reasons why this algorithm has been so successful. Finally, in chapter 8 we describe an application of machine learning to video games, where computational efficiency is tied to software and hardware engineering constraints which, although often ignored in research papers, are ubiquitous in practice.

Malgré des progrès constants en termes de capacité de calcul, mémoire et quantité de données disponibles, les algorithmes d'apprentissage machine doivent se montrer efficaces dans l'utilisation de ces ressources. La minimisation des coûts est évidemment un facteur important, mais une autre motivation est la recherche de mécanismes d'apprentissage capables de reproduire le comportement d'êtres intelligents. Cette thèse aborde le problème de l'efficacité à travers plusieurs articles traitant d'algorithmes d'apprentissage variés : ce problème est vu non seulement du point de vue de l'efficacité computationnelle (temps de calcul et mémoire utilisés), mais aussi de celui de l'efficacité statistique (nombre d'exemples requis pour accomplir une tâche donnée). Une première contribution apportée par cette thèse est la mise en lumière d'inefficacités statistiques dans des algorithmes existants. Nous montrons ainsi que les arbres de décision généralisent mal pour certains types de tâches (chapitre 3), de même que les algorithmes classiques d'apprentissage semi-supervisé à base de graphe (chapitre 5), chacun étant affecté par une forme particulière de la malédiction de la dimensionalité. Pour une certaine classe de réseaux de neurones, appelés réseaux sommes-produits, nous montrons qu'il peut être exponentiellement moins efficace de représenter certaines fonctions par des réseaux à une seule couche cachée, comparé à des réseaux profonds (chapitre 4). Nos analyses permettent de mieux comprendre certains problèmes intrinsèques liés à ces algorithmes, et d'orienter la recherche dans des directions qui pourraient permettre de les résoudre. Nous identifions également des inefficacités computationnelles dans les algorithmes d'apprentissage semi-supervisé à base de graphe (chapitre 5), et dans l'apprentissage de mélanges de Gaussiennes en présence de valeurs manquantes (chapitre 6). Dans les deux cas, nous proposons de nouveaux algorithmes capables de traiter des ensembles de données significativement plus grands. Les deux derniers chapitres traitent de l'efficacité computationnelle sous un angle différent. Dans le chapitre 7, nous analysons de manière théorique un algorithme existant pour l'apprentissage efficace dans les machines de Boltzmann restreintes (la divergence contrastive), afin de mieux comprendre les raisons qui expliquent le succès de cet algorithme. Finalement, dans le chapitre 8 nous présentons une application de l'apprentissage machine dans le domaine des jeux vidéo, pour laquelle le problème de l'efficacité computationnelle est relié à des considérations d'ingénierie logicielle et matérielle, souvent ignorées en recherche mais ô combien importantes en pratique. / Despite constant progress in terms of available computational power, memory and amount of data, machine learning algorithms need to be efficient in how they use them. Although minimizing cost is an obvious major concern, another motivation is to attempt to design algorithms that can learn as efficiently as intelligent species. This thesis tackles the problem of efficient learning through various papers dealing with a wide range of machine learning algorithms: this topic is seen both from the point of view of computational efficiency (processing power and memory required by the algorithms) and of statistical efficiency (n umber of samples necessary to solve a given learning task).The first contribution of this thesis is in shedding light on various statistical inefficiencies in existing algorithms. Indeed, we show that decision trees do not generalize well on tasks with some particular properties (chapter 3), and that a similar flaw affects typical graph-based semi-supervised learning algorithms (chapter 5). This flaw is a form of curse of dimensionality that is specific to each of these algorithms. For a subclass of neural networks, called sum-product networks, we prove that using networks with a single hidden layer can be exponentially less efficient than when using deep networks (chapter 4). Our analyses help better understand some inherent flaws found in these algorithms, and steer research towards approaches that may potentially overcome them. We also exhibit computational inefficiencies in popular graph-based semi-supervised learning algorithms (chapter 5) as well as in the learning of mixtures of Gaussians with missing data (chapter 6). In both cases we propose new algorithms that make it possible to scale to much larger datasets. The last two chapters also deal with computational efficiency, but in different ways. Chapter 7 presents a new view on the contrastive divergence algorithm (which has been used for efficient training of restricted Boltzmann machines). It provides additional insight on the reasons why this algorithm has been so successful. Finally, in chapter 8 we describe an application of machine learning to video games, where computational efficiency is tied to software and hardware engineering constraints which, although often ignored in research papers, are ubiquitous in practice.