Machine learning has grown in popularity to help assign resources and make decisions about users, which can result in discrimination. This includes hiring markets, where employers have increasingly been interested in using automated tools to help hire candidates. In response, there has been significant effort to understand and mitigate the sources of discrimination in these tools. However, previous work has largely assumed that discrimination, in any area of ML, is the result of some initial unequal distribution of resources across groups: One group is on average less qualified, there is less training data for one group, or the classifier is less accurate on one group, etc. However, recent work have suggested that there are other sources of discrimination, such as relational inequality, that are notably non-distributional. First, we show consensus in strategy choice is a non-distributional source of inequality at equilibrium in games: We provide subgame perfect equilibria in a simple sequential model of a hiring market with Rubinstein-style bargaining between firms and candidates that exhibits asymmetric wages resulting from differences in agents' threat strategies during bargaining. Second, we give an initial analysis of how agents could learn such strategies via convergence of an online learning algorithm to asymmetric equilibria. Ultimately, this work motivates the further study of endogenous, possibly non-distributional, mechanisms of inequality in ML.

Algorithms and other formal models purportedly incorporating human values like fairness have grown increasingly popular in computer science. In response to sociotechnical challenges in the use of these models, designers and researchers have taken widely divergent positions on how formal models incorporating aspects of human values should be used: encouraging their use, moving away from them, or ignoring the normative consequences altogether. In this paper, we seek to resolve these divergent positions by identifying the main conceptual limits of formal modeling, and develop four reflexive values--value fidelity, appropriate accuracy, value legibility, and value contestation--vital for incorporating human values adequately into formal models. We then provide a brief methodology for reflexively designing formal models incorporating human values.

In this work, we study how to use sampling to speed up mechanisms for answering adaptive queries into datasets without reducing the accuracy of those mechanisms. This is important to do when both the datasets and the number of queries asked are very large. In particular, we describe a mechanism that provides a polynomial speed-up per query over previous mechanisms, without needing to increase the total amount of data required to maintain the same generalization error as before. We prove that this speed-up holds for arbitrary statistical queries. We also provide an even faster method for achieving statistically-meaningful responses wherein the mechanism is only allowed to see a constant number of samples from the data per query. Finally, we show that our general results yield a simple, fast, and unified approach for adaptively optimizing convex and strongly convex functions over a dataset.

In the problem of learning a class ratio from unlabeled data, which we call CR learning, the training data is unlabeled, and only the ratios, or proportions, of examples receiving each label are given. The goal is to learn a hypothesis that predicts the proportions of labels on the distribution underlying the sample. This model of learning is applicable to a wide variety of settings, including predicting the number of votes for candidates in political elections from polls. In this paper, we formally define this class and resolve foundational questions regarding the computational complexity of CR learning and characterize its relationship to PAC learning. Among our results, we show, perhaps surprisingly, that for finite VC classes what can be efficiently CR learned is a strict subset of what can be learned efficiently in PAC, under standard complexity assumptions. We also show that there exist classes of functions whose CR learnability is independent of ZFC, the standard set theoretic axioms. This implies that CR learning cannot be easily characterized (like PAC by VC dimension).

The study of influence maximization in social networks has largely ignored disparate effects these algorithms might have on the individuals contained in the social network. Individuals may place a high value on receiving information, e.g. job openings or advertisements for loans. While well-connected individuals at the center of the network are likely to receive the information that is being distributed through the network, poorly connected individuals are systematically less likely to receive the information, producing a gap in access to the information between individuals. In this work, we study how best to spread information in a social network while minimizing this access gap.

Learning automata is a foundational problem in computational learning theory. However, even efficiently learning random DFAs is hard. A natural restriction of this problem is to consider learning random DFAs under the uniform distribution. To date, this problem has no non-trivial lower bounds nor algorithms faster than brute force. In this note, we propose a method to find faster algorithms for this problem. We reduce the learning problem to a conjecture about meeting times of random walks over random DFAs, which may be of independent interest to prove.

In the problem of learning with label proportions (also known as the problem of estimating class ratios), the training data is unlabeled, and only the proportions of examples receiving each label are given. The goal is to learn a hypothesis that predicts the proportions of labels on the distribution underlying the sample. This model of learning is useful in a wide variety of settings, including predicting the number of votes for candidates in political elections from polls. In this paper, we resolve foundational questions regarding the computational complexity of learning in this setting. We formalize a simple version of the setting, and we compare the computational complexity of learning in this model to classical PAC learning. Perhaps surprisingly, we show that what can be learned efficiently in this model is a strict subset of what may be leaned efficiently in PAC, under standard complexity assumptions. We give a characterization in terms of VC dimension, and we show that there are non-trivial problems in this model that can be efficiently learned. We also give an algorithm that demonstrates the feasibility of learning under well-behaved distributions.

We study three classical machine learning algorithms in the context of algorithmic fairness: adaptive boosting, support vector machines, and logistic regression. Our goal is to maintain the high accuracy of these learning algorithms while reducing the degree to which they discriminate against individuals because of their membership in a protected group. Our first contribution is a method for achieving fairness by shifting the decision boundary for the protected group. The method is based on the theory of margins for boosting. Our method performs comparably to or outperforms previous algorithms in the fairness literature in terms of accuracy and low discrimination, while simultaneously allowing for a fast and transparent quantification of the trade-off between bias and error. Our second contribution addresses the shortcomings of the bias-error trade-off studied in most of the algorithmic fairness literature. We demonstrate that even hopelessly naive modifications of a biased algorithm, which cannot be reasonably said to be fair, can still achieve low bias and high accuracy. To help to distinguish between these naive algorithms and more sensible algorithms we propose a new measure of fairness, called resilience to random bias (RRB). We demonstrate that RRB distinguishes well between our naive and sensible fairness algorithms. RRB together with bias and accuracy provides a more complete picture of the fairness of an algorithm.

In this work, we consider a classification problem of 14 physical activities using a body sensor network (BSN) consisting of 14 tri-axial accelerometers. We use a tree-based classifier, and develop a feature selection algorithm based on mutual information to find the relevant features at every internal node of the tree. We evaluate our algorithm on 31 features per accelerometer (total of 434), and we present the results on 8 subjects with a 96% average accuracy.