Recent work in fairness in machine learning has proposed adjusting for fairness by equalizing accuracy metrics across groups and has also studied how datasets affected by historical prejudices may lead to unfair decision policies. We connect these lines of work and study the residual unfairness that arises when a fairness-adjusted predictor is not actually fair on the target population due to systematic censoring of training data by existing biased policies. This scenario is particularly common in the same applications where fairness is a concern. We characterize theoretically the impact of such censoring on standard fairness metrics for binary classifiers and provide criteria for when residual unfairness may or may not appear. We prove that, under certain conditions, fairness-adjusted classifiers will in fact induce residual unfairness that perpetuates the same injustices, against the same groups, that biased the data to begin with, thus showing that even state-of-the-art fair machine learning can have a "bias in, bias out" property. When certain benchmark data is available, we show how sample reweighting can estimate and adjust fairness metrics while accounting for censoring. We use this to study the case of Stop, Question, and Frisk (SQF) and demonstrate that attempting to adjust for fairness perpetuates the same injustices that the policy is infamous for.

Summary: There is a great hue and cry about the danger of bias in our predictive models when applied to high significance events like who gets a loan, insurance, a good school assignment, or bail. It's not as simple as it seems and here we try to take a more nuanced look. The result is not as threatening as many headlines make it seem. Is social bias in our models a threat to equal and fair treatment? There's even an entire conference dedicated to the topic: the conference on Fairness, Accountability, and Transparency (FAT* – it's their acronym, I didn't make this up) now in its fifth year.

Automated surveillance is going to become increasingly common as companies and researchers find new ways to use machine learning to analyze live video footage. A new project from scientists in the UK and India shows one possible use for this technology: identifying violent behavior in crowds with the help of camera-equipped drones. In a paper titled "Eye in the Sky," the researchers describe their system. It uses a simple Parrot AR quadcopter (which costs around $200) to transmit video footage over a mobile internet connection for real-time analysis. An algorithm trained using deep learning estimates the poses of humans in the video and matches them to postures the researchers have designated as "violent."

The 2017 ASSISTments Data Mining competition aims to use data from a longitudinal study for predicting a brand-new outcome of students which had never been studied before by the educational data mining research community. Specifically, it facilitates research in developing predictive models that predict whether the first job of a student out of college belongs to a STEM (the acronym for science, technology, engineering, and mathematics) field. This is based on the student's learning history on the ASSISTments blended learning platform in the form of extensive clickstream data gathered during the middle school years. To tackle this challenge, we first estimate the expected knowledge state of students with respect to different mathematical skills using a deep knowledge tracing (DKT) model and an enhanced DKT (DKT+) model. We then combine the features corresponding to the DKT/DKT+ expected knowledge state with other features extracted directly from the student profile in the dataset to train several machine learning models for the STEM/non-STEM job prediction. Our experiments show that models trained with the combined features generally perform better than the models trained with the student profile alone. Detailed analysis of the student's knowledge state reveals that, when compared with non-STEM students, STEM students generally show a higher mastery level and a higher learning gain in mathematics.

"Machine learning" is a computer science discipline that refers to the ability for machines to learn with data and carry out tasks that would typically require human intelligence. The technology is growing quickly: according to Gartner, more than half of data and analytics services will be performed by machines rather than human beings by 2022, which is 10 percent more than today. The emergence of machine learning and its implementation into consumer facing applications coincides conveniently with today's real-time economy. Machine learning drives a decrease in fraud before it impacts the victim, just as our society has become as impatient as ever. In fact, more than 60 percent of people increasingly feel that waiting for something that should happen instantaneously impacts their perception of the underlying brand -- which is especially true when it comes to identity or financial fraud.

This Ph.D. thesis deals with the optimization of several renewable energy resources development as well as the improvement of facilities management in oceanic engineering and airports, using computational hybrid methods belonging to AI to this end. Energy is essential to our society in order to ensure a good quality of life. This means that predictions over the characteristics on which renewable energies depend are necessary, in order to know the amount of energy that will be obtained at any time. The second topic tackled in this thesis is related to the basic parameters that influence in different marine activities and airports, whose knowledge is necessary to develop a proper facilities management in these environments. Within this work, a study of the state-of-the-art Machine Learning have been performed to solve the problems associated with the topics above-mentioned, and several contributions have been proposed: One of the pillars of this work is focused on the estimation of the most important parameters in the exploitation of renewable resources. The second contribution of this thesis is related to feature selection problems. The proposed methodologies are applied to multiple problems: the prediction of $H_s$, relevant for marine energy applications and marine activities, the estimation of WPREs, undesirable variations in the electric power produced by a wind farm, the prediction of global solar radiation in areas from Spain and Australia, really important in terms of solar energy, and the prediction of low-visibility events at airports. All of these practical issues are developed with the consequent previous data analysis, normally, in terms of meteorological variables.

We propose a new semiparametric approach to binary classification that exploits the modeling flexibility of sparse graphical models. Specifically, we assume that each class can be represented by a forest-structured graphical model. Under this assumption, the optimal classifier is linear in the log of the one- and two-dimensional marginal densities. Our proposed procedure non-parametrically estimates the univariate and bivariate marginal densities, maps each sample to the logarithm of these estimated densities and constructs a linear SVM in the transformed space. We prove convergence of the resulting classifier to an oracle SVM classifier and give finite sample bounds on its excess risk. Experiments with simulated and real data indicate that the resulting classifier is competitive with several popular methods across a range of applications.

We present MRPC, an R package that learns causal graphs with improved accuracy over existing packages, such as pcalg and bnlearn. Our algorithm builds on the powerful PC algorithm, the canonical algorithm in computer science for learning directed acyclic graphs. The improvement in accuracy results from online control of the false discovery rate (FDR) that reduces false positive edges, a more accurate approach to identifying v-structures (i.e., $T_1 \rightarrow T_2 \leftarrow T_3$), and robust estimation of the correlation matrix among nodes. For genomic data that contain genotypes and gene expression for each sample, MRPC incorporates the principle of Mendelian randomization to orient the edges. Our package can be applied to continuous and discrete data.

Given a binary prediction problem, which performance metric should the classifier optimize? We address this question by formalizing the problem of metric elicitation. In particular, we focus on eliciting binary performance metrics from pairwise preferences, where users provide relative feedback for pairs of classifiers. By exploiting key properties of the space of confusion matrices, we obtain provably query efficient algorithms for eliciting linear and linear-fractional metrics. We further show that our method is robust to feedback and finite sample noise.

Graphs are widely used as a natural framework that captures interactions between individual elements represented as nodes in a graph. In medical applications, specifically, nodes can represent individuals within a potentially large population (patients or healthy controls) accompanied by a set of features, while the graph edges incorporate associations between subjects in an intuitive manner. This representation allows to incorporate the wealth of imaging and non-imaging information as well as individual subject features simultaneously in disease classification tasks. Previous graph-based approaches for supervised or unsupervised learning in the context of disease prediction solely focus on pairwise similarities between subjects, disregarding individual characteristics and features, or rather rely on subject-specific imaging feature vectors and fail to model interactions between them. In this paper, we present a thorough evaluation of a generic framework that leverages both imaging and non-imaging information and can be used for brain analysis in large populations. This framework exploits Graph Convolutional Networks (GCNs) and involves representing populations as a sparse graph, where its nodes are associated with imaging-based feature vectors, while phenotypic information is integrated as edge weights. The extensive evaluation explores the effect of each individual component of this framework on disease prediction performance and further compares it to different baselines. The framework performance is tested on two large datasets with diverse underlying data, ABIDE and ADNI, for the prediction of Autism Spectrum Disorder and conversion to Alzheimer's disease, respectively. Our analysis shows that our novel framework can improve over state-of-the-art results on both databases, with 70.4% classification accuracy for ABIDE and 80.0% for ADNI.