We study a mean-field spike and slab variational Bayes (VB) approximation to Bayesian model selection priors in sparse high-dimensional linear regression. Under compatibility conditions on the design matrix, oracle inequalities are derived for the mean-field VB approximation, implying that it converges to the sparse truth at the optimal rate and gives optimal prediction of the response vector. The empirical performance of our algorithm is studied, showing that it works comparably well as other state-of-the-art Bayesian variable selection methods. We also numerically demonstrate that the widely used coordinate-ascent variational inference (CAVI) algorithm can be highly sensitive to the parameter updating order, leading to potentially poor performance. To mitigate this, we propose a novel prioritized updating scheme that uses a data-driven updating order and performs better in simulations.

Ideal point estimation and dimensionality reduction have long been utilized to simplify and cluster complex, high-dimensional political data (e.g., roll-call votes and surveys) for use in analysis and visualization. These methods often work by finding the directions or principal components (PCs) on which either the data varies the most or respondents make the fewest decision errors. However, these PCs, which usually reflect the left-right political spectrum, are sometimes uninformative in explaining significant differences in the distribution of the data (e.g., how to categorize a set of highly-moderate voters). To tackle this issue, we adopt an emerging analysis approach, called contrastive learning. Contrastive learning-e.g., contrastive principal component analysis (cPCA)-works by first splitting the data by predefined groups, and then deriving PCs on which the target group varies the most but the background group varies the least. As a result, cPCA can often find `hidden' patterns, such as subgroups within the target group, which PCA cannot reveal when some variables are the dominant source of variations across the groups. We contribute to the field of contrastive learning by extending it to multiple correspondence analysis (MCA) to enable an analysis of data often encountered by social scientists---namely binary, ordinal, and nominal variables. We demonstrate the utility of contrastive MCA (cMCA) by analyzing three different surveys: The 2015 Cooperative Congressional Election Study, 2012 UTokyo-Asahi Elite Survey, and 2018 European Social Survey. Our results suggest that, first, for the cases when ordinary MCA depicts differences between groups, cMCA can further identify characteristics that divide the target group; second, for the cases when MCA does not show clear differences, cMCA can successfully identify meaningful directions and subgroups, which traditional methods overlook.

The impact of gradient noise on training deep models is widely acknowledged but not well understood. In this context, we study the distribution of gradients during training. We introduce a method, Gradient Clustering, to minimize the variance of average mini-batch gradient with stratified sampling. We prove that the variance of average mini-batch gradient is minimized if the elements are sampled from a weighted clustering in the gradient space. We measure the gradient variance on common deep learning benchmarks and observe that, contrary to common assumptions, gradient variance increases during training, and smaller learning rates coincide with higher variance. In addition, we introduce normalized gradient variance as a statistic that better correlates with the speed of convergence compared to gradient variance.

As machine learning systems become increasingly entangled with human decision-making processes, it is of social, ethical, and economic importance to be able to understand and explain the limits of what can be said about such technologies. We need to be able to formulate meaningful questions about the behavior or performance of learning systems that are accessible to a diverse audience, and we must have the tools required to provide unambiguous answers to these questions. From the viewpoint of responsible and sustainable system design, the tools used to answer questions about the learning system are arguably even more important than the algorithms underlying the learning system itself. In this context, there is an important line of research related to conformal prediction, a general-purpose methodology for attaching high-probability guarantees to data-driven predictors [25, 16, 9, 8]. For example, if the user specifies a confidence level of say 90% in advance, questions such as "I've trained a classifier. Given a new instance, what labels seem most likely?"

We propose new tools for policy-makers to use when assessing and correcting fairness and bias in AI algorithms. The three tools are: - A new definition of fairness called "controlled fairness" with respect to choices of protected features and filters. The definition provides a simple test of fairness of an algorithm with respect to a dataset. This notion of fairness is suitable in cases where fairness is prioritized over accuracy, such as in cases where there is no "ground truth" data, only data labeled with past decisions (which may have been biased). - Algorithms for retraining a given classifier to achieve "controlled fairness" with respect to a choice of features and filters. Two algorithms are presented, implemented and tested. These algorithms require training two different models in two stages. We experiment with combinations of various types of models for the first and second stage and report on which combinations perform best in terms of fairness and accuracy. - Algorithms for adjusting model parameters to achieve a notion of fairness called "classification parity". This notion of fairness is suitable in cases where accuracy is prioritized. Two algorithms are presented, one which assumes that protected features are accessible to the model during testing, and one which assumes protected features are not accessible during testing. We evaluate our tools on three different publicly available datasets. We find that the tools are useful for understanding various dimensions of bias, and that in practice the algorithms are effective in starkly reducing a given observed bias when tested on new data.

We build a Bayesian contextual classification model using an optimistic score ratio for robust binary classification when there is limited information on the class-conditional, or contextual, distribution. The optimistic score searches for the distribution that is most plausible to explain the observed outcomes in the testing sample among all distributions belonging to the contextual ambiguity set which is prescribed using a limited structural constraint on the mean vector and the covariance matrix of the underlying contextual distribution. We show that the Bayesian classifier using the optimistic score ratio is conceptually attractive, delivers solid statistical guarantees and is computationally tractable. We showcase the power of the proposed optimistic score ratio classifier on both synthetic and empirical data.

We introduce online probabilistic label trees (OPLTs), an algorithm that trains a label tree classifier in a fully online manner, without any prior knowledge about the number of training instances, their features and labels. OPLTs are characterized by low time and space complexity as well as strong theoretical guarantees. They can be used for online multi-label and multi-class classification, including the very challenging scenarios of one- or few-shot learning. We demonstrate the attractiveness of OPLTs in a wide empirical study on several instances of the tasks mentioned above.

Generalized fiducial inference(GFI) Hannig et al.(2016), a modern reincarnation of R. A. Fisher's fiducial inference(Fisher, 1930), provides inferentially meaningful probability statements about subsets of parameter space without the need for subjective prior information. GFI specifies a generalized fiducial distribution (GFD) by defining a data-dependent measure on the parameter space through an inverse of a data-generating algorithm (see Sections 2). Data-generating algorithm plays the role of a model and is sometimes called data-generating equation or data-generating function. With GFD as a distribution estimator for the fixed parameter, we can further define approximate confidence (fiducial) sets which are often shown in simulation to have very desired properties. Given the data-generating algorithm and the corresponding density of the GFD, one could form point estimate and asymptotic confidence sets similarly as with a Bayesian posterior density. Standard MCMC-type sampling techniques have already been successfully implemented in many situations; see Hannig et al. (2016) and the references therein.

In the context of few-shot learning, one cannot measure the generalization ability of a trained classifier using validation sets, due to the small number of labeled samples. In this paper, we are interested in finding alternatives to answer the question: is my classifier generalizing well to previously unseen data? We first analyze the reasons for the variability of generalization performances. We then investigate the case of using transfer-based solutions, and consider three settings: i) supervised where we only have access to a few labeled samples, ii) semi-supervised where we have access to both a few labeled samples and a set of unlabeled samples and iii) unsupervised where we only have access to unlabeled samples. For each setting, we propose reasonable measures that we empirically demonstrate to be correlated with the generalization ability of considered classifiers. We also show that these simple measures can be used to predict generalization up to a certain confidence. We conduct our experiments on standard few-shot vision datasets.

The success of adversarial formulations in machine learning has brought renewed motivation for smooth games. In this work, we focus on the class of stochastic Hamiltonian methods and provide the first convergence guarantees for certain classes of stochastic smooth games. We propose a novel unbiased estimator for the stochastic Hamiltonian gradient descent (SHGD) and highlight its benefits. Using tools from the optimization literature we show that SHGD converges linearly to the neighbourhood of a stationary point. To guarantee convergence to the exact solution, we analyze SHGD with a decreasing step-size and we also present the first stochastic variance reduced Hamiltonian method. Our results provide the first global non-asymptotic last-iterate convergence guarantees for the class of stochastic unconstrained bilinear games and for the more general class of stochastic games that satisfy a "sufficiently bilinear" condition, notably including some non-convex non-concave problems. We supplement our analysis with experiments on stochastic bilinear and sufficiently bilinear games, where our theory is shown to be tight, and on simple adversarial machine learning formulations.