The goal of a regression problem in machine learning is to find the value of a function that can accurately predict the data pattern. Similarly, a classification problem involves finding the value of the function that can accurately classify the different classes of data. The accuracy of the model is determined on the basis of how well the model predicts the output values, given the input values. Here, we will be discussing one such metric used in iteratively calibrating the accuracy of the model, known as the cost function. Before answering the question of how does the model learn, it is important to know what does the model actually learn?

Understanding how privacy of an individual degrades as the number of analyses using their data grows is of paramount importance in privacy-preserving data analysis. On one hand, this allows individuals to participate in multiple disjoint statistical analyses, all the while knowing that their privacy cannot be compromised by aggregating the resulting reports. On the other hand, this feature is crucial for privacy-preserving algorithm design -- instead of having to reason about the privacy properties of a complex algorithm, it allows reasoning about the privacy of the subroutines that make up the final algorithm. For differential privacy [11], this accounting of privacy losses is typically done using composition theorems. Importantly, given that statistical analyses often rely on the outputs of previous analyses, and that algorithmic subroutines feed into one another, the composition theorems need to be adaptive, namely, allow the choice of which algorithm to run next to depend on the outputs of all previous computations. For example, in gradient descent, the computation of the gradient depends on the value of the current iterate, which itself is the output of the previous steps of the algorithm. Given the central role that adaptive composition theorems play for differentially private data analysis, they have been investigated in numerous works (e.g.

This thesis focuses on the research and development of the Hemodynamic Tissue Signature (HTS) method: an unsupervised machine learning approach to describe the vascular heterogeneity of glioblastomas by means of perfusion MRI analysis. The HTS builds on the concept of habitats. An habitat is defined as a sub-region of the lesion with a particular MRI profile describing a specific physiological behavior. The HTS method delineates four habitats within the glioblastoma: the High Angiogenic Tumor (HAT) habitat, as the most perfused region of the enhancing tumor; the Low Angiogenic Tumor (LAT) habitat, as the region of the enhancing tumor with a lower angiogenic profile; the potentially Infiltrated Peripheral Edema (IPE) habitat, as the non-enhancing region adjacent to the tumor with elevated perfusion indexes; and the Vasogenic Peripheral Edema (VPE) habitat, as the remaining edema of the lesion with the lowest perfusion profile. The results of this thesis have been published in ten scientific contributions, including top-ranked journals and conferences in the areas of Medical Informatics, Statistics and Probability, Radiology & Nuclear Medicine, Machine Learning and Data Mining and Biomedical Engineering. An industrial patent registered in Spain (ES201431289A), Europe (EP3190542A1) and EEUU (US20170287133A1) was also issued, summarizing the efforts of the thesis to generate tangible assets besides the academic revenue obtained from research publications. Finally, the methods, technologies and original ideas conceived in this thesis led to the foundation of ONCOANALYTICS CDX, a company framed into the business model of companion diagnostics for pharmaceutical compounds, thought as a vehicle to facilitate the industrialization of the ONCOhabitats technology.

As a technology ML is oblivious to societal good or bad, and thus, the field of fair machine learning has stepped up to propose multiple mathematical definitions, algorithms, and systems to ensure different notions of fairness in ML applications. Given the multitude of propositions, it has become imperative to formally verify the fairness metrics satisfied by different algorithms on different datasets. In this paper, we propose a \textit{stochastic satisfiability} (SSAT) framework, Justicia, that formally verifies different fairness measures of supervised learning algorithms with respect to the underlying data distribution. We instantiate Justicia on multiple classification and bias mitigation algorithms, and datasets to verify different fairness metrics, such as disparate impact, statistical parity, and equalized odds. Justicia is scalable, accurate, and operates on non-Boolean and compound sensitive attributes unlike existing distribution-based verifiers, such as FairSquare and VeriFair. Being distribution-based by design, Justicia is more robust than the verifiers, such as AIF360, that operate on specific test samples. We also theoretically bound the finite-sample error of the verified fairness measure.

This papers studies how competition affects machine learning (ML) predictors. As ML becomes more ubiquitous, it is often deployed by companies to compete over customers. For example, digital platforms like Yelp use ML to predict user preference and make recommendations. A service that is more often queried by users, perhaps because it more accurately anticipates user preferences, is also more likely to obtain additional user data (e.g. in the form of a Yelp review). Thus, competing predictors cause feedback loops whereby a predictor's performance impacts what training data it receives and biases its predictions over time. We introduce a flexible model of competing ML predictors that enables both rapid experimentation and theoretical tractability. We show with empirical and mathematical analysis that competition causes predictors to specialize for specific sub-populations at the cost of worse performance over the general population. We further analyze the impact of predictor specialization on the overall prediction quality experienced by users. We show that having too few or too many competing predictors in a market can hurt the overall prediction quality. Our theory is complemented by experiments on several real datasets using popular learning algorithms, such as neural networks and nearest neighbor methods.

Having a large number of covariates can have a negative impact on the quality of causal effect estimation since confounding adjustment becomes unreliable when the number of covariates is large relative to the samples available. Propensity score is a common way to deal with a large covariate set, but the accuracy of propensity score estimation (normally done by logistic regression) is also challenged by large number of covariates. In this paper, we prove that a large covariate set can be reduced to a lower dimensional representation which captures the complete information for adjustment in causal effect estimation. The theoretical result enables effective data-driven algorithms for causal effect estimation. We develop an algorithm which employs a supervised kernel dimension reduction method to search for a lower dimensional representation for the original covariates, and then utilizes nearest neighbor matching in the reduced covariate space to impute the counterfactual outcomes to avoid large-sized covariate set problem. The proposed algorithm is evaluated on two semi-synthetic and three real-world datasets and the results have demonstrated the effectiveness of the algorithm.

Neural networks are a general framework for differentiable optimization which includes many other machine learning approaches as special cases. In this paper we build a category-theoretic formalism around a neural network system called CycleGAN. CycleGAN is a general approach to unpaired image-to-image translation that has been getting attention in the recent years. Inspired by categorical database systems, we show that CycleGAN is a "schema", i.e. a specific category presented by generators and relations, whose specific parameter instantiations are just set-valued functors on this schema. We show that enforcing cycle-consistencies amounts to enforcing composition invariants in this category. We generalize the learning procedure to arbitrary such categories and show a special class of functors, rather than functions, can be learned using gradient descent. Using this framework we design a novel neural network system capable of learning to insert and delete objects from images without paired data. We qualitatively evaluate the system on the CelebA dataset and obtain promising results.

A straightforward application of semi-supervised machine learning to the problem of treatment effect estimation would be to consider data as "unlabeled" if treatment assignment and covariates are observed but outcomes are unobserved. According to this formulation, large unlabeled data sets could be used to estimate a high dimensional propensity function and causal inference using a much smaller labeled data set could proceed via weighted estimators using the learned propensity scores. In the limiting case of infinite unlabeled data, one may estimate the high dimensional propensity function exactly. However, longstanding advice in the causal inference community suggests that estimated propensity scores (from labeled data alone) are actually preferable to true propensity scores, implying that the unlabeled data is actually useless in this context. In this paper we examine this paradox and propose a simple procedure that reconciles the strong intuition that a known propensity functions should be useful for estimating treatment effects with the previous literature suggesting otherwise. Further, simulation studies suggest that direct regression may be preferable to inverse-propensity weight estimators in many circumstances.

Community detection is a fundamental problem in the analysis of complex networks. It is the analogue of clustering in network data mining. Within community detection methods, hierarchical algorithms are popular. However, their iterative nature and the need to recompute the structural properties used to split the network (i.e. edge betweenness in Girvan and Newman's algorithm), make them unsuitable for large network data sets. In this paper, we study how local structural network properties can be used as proxies to improve the efficiency of hierarchical community detection while, at the same time, achieving competitive results in terms of modularity. In particular, we study the potential use of the structural properties commonly used to perform local link prediction, a supervised learning problem where community structure is relevant, as nodes are prone to establish new links with other nodes within their communities. In addition, we check the performance impact of network pruning heuristics as an ancillary tactic to make hierarchical community detection more efficient

This master's thesis work has the objective of performing an analysis of the methods for detecting communities in networks. As an initial part, I study of the main features of graph theory and communities, as well as common measures in this problem. Subsequently, I was performed a review of the main methods of detecting communities, developing a classification, taking into account its characteristics and computational complexity for the detection of strengths and weaknesses in the methods, as well as later works. Then, study the problem of classification of a clustering method, this in order to evaluate the quality of the communities detected by analyzing different measures. Finally conclusions are elaborated and possible lines of work that can be derived.