skip to main content

Title: DENOUNCER: detection of unfairness in classifiers
The use of automated data-driven tools for decision-making has gained popularity in recent years. At the same time, the reported cases of algorithmic bias and discrimination increase as well, which in turn lead to an extensive study of algorithmic fairness. Numerous notions of fairness have been proposed, designed to capture different scenarios. These measures typically refer to a "protected group" in the data, defined using values of some sensitive attributes. Confirming whether a fairness definition holds for a given group is a simple task, but detecting groups that are treated unfairly by the algorithm may be computationally prohibitive as the number of possible groups is combinatorial. We present a method for detecting such groups efficiently for various fairness definitions. Our solution is implemented in a system called DENOUNCER, an interactive system that allows users to explore different fairness measures of a (trained) classifier for a given test data. We propose to demonstrate the usefulness of DENOUNCER using real-life data and illustrate the effectiveness of our method.
; ;
Award ID(s):
1741022 1934565 2106176
Publication Date:
Journal Name:
Proceedings of the VLDB Endowment
Page Range or eLocation-ID:
2719 to 2722
Sponsoring Org:
National Science Foundation
More Like this
  1. Welfare measures overall utility across a population, whereas malfare measures overall disutility, and the social planner’s problem can be cast either as maximizing the former or minimizing the latter. We show novel bounds on the expectations and tail probabilities of estimators of welfare, malfare, and regret of per-group (dis)utility values, where estimates are made from a finite sample drawn from each group. In particular, we consider estimating these quantities for individual functions (e.g., allocations or classifiers) with standard probabilistic bounds, and optimizing and bounding generalization error over hypothesis classes (i.e., we quantify overfitting) using Rademacher averages. We then study algorithmic fairness through the lens of sample complexity, finding that because marginalized or minority groups are often understudied, and fewer data are therefore available, the social planner is more likely to overfit to these groups, thus even models that seem fair in training can be systematically biased against such groups. We argue that this effect can be mitigated by ensuring sufficient sample sizes for each group, and our sample complexity analysis characterizes these sample sizes. Motivated by these conclusions, we present progressive sampling algorithms to efficiently optimize various fairness objectives.
  2. Recently there has been a growing interest in fairness-aware recommender systems including fairness in providing consistent performance across different users or groups of users. A recommender system could be considered unfair if the recommendations do not fairly represent the tastes of a certain group of users while other groups receive recommendations that are consistent with their preferences. In this paper, we use a metric called miscalibration for measuring how a recommendation algorithm is responsive to users’ true preferences and we consider how various algorithms may result in different degrees of miscalibration for different users. In particular, we conjecture that popularity bias which is a well-known phenomenon in recommendation is one important factor leading to miscalibration in recommendation. Our experimental results using two real-world datasets show that there is a connection between how different user groups are affected by algorithmic popularity bias and their level of interest in popular items. Moreover, we show that the more a group is affected by the algorithmic popularity bias, the more their recommendations are miscalibrated.
  3. Fair consensus building combines the preferences of multiple rankers into a single consensus ranking, while ensuring any group defined by a protected attribute (such as race or gender) is not disadvantaged compared to other groups. Manually generating a fair consensus ranking is time-consuming and impractical- even for a fairly small number of candidates. While algorithmic approaches for auditing and generating fair consensus rankings have been developed, these have not been operationalized in interactive systems. To bridge this gap, we introduce FairFuse, a visualization system for generating, analyzing, and auditing fair consensus rankings. We construct a data model which includes base rankings entered by rankers, augmented with measures of group fairness, and algorithms for generating consensus rankings with varying degrees of fairness. We design novel visualizations that encode these measures in a parallel-coordinates style rank visualization, with interactions for generating and exploring fair consensus rankings. We describe use cases in which FairFuse supports a decision-maker in ranking scenarios in which fairness is important, and discuss emerging challenges for future efforts supporting fairness-oriented rank analysis. Code and demo videos available at
  4. Settings such as lending and policing can be modeled by a centralized agent allocating a scarce resource (e.g. loans or police officers) amongst several groups, in order to maximize some objective (e.g. loans given that are repaid, or criminals that are apprehended). Often in such problems fairness is also a concern. One natural notion of fairness, based on general principles of equality of opportunity, asks that conditional on an individual being a candidate for the resource in question, the probability of actually receiving it is approximately independent of the individual’s group. For example, in lending this would mean that equally creditworthy individuals in different racial groups have roughly equal chances of receiving a loan. In policing it would mean that two individuals committing the same crime in different districts would have roughly equal chances of being arrested. In this paper, we formalize this general notion of fairness for allocation problems and investigate its algorithmic consequences. Our main technical results include an efficient learning algorithm that converges to an optimal fair allocation even when the allocator does not know the frequency of candidates (i.e. creditworthy individuals or criminals) in each group. This algorithm operates in a censored feedback model in whichmore »only the number of candidates who received the resource in a given allocation can be observed, rather than the true number of candidates in each group. This models the fact that we do not learn the creditworthiness of individuals we do not give loans to and do not learn about crimes committed if the police presence in a district is low.« less
  5. Algorithmic decisions made by machine learning models in high-stakes domains may have lasting impacts over time. However, naive applications of standard fairness criterion in static settings over temporal domains may lead to delayed and adverse effects. To understand the dynamics of performance disparity, we study a fairness problem in Markov decision processes (MDPs). Specifically, we propose return parity, a fairness notion that requires MDPs from different demographic groups that share the same state and action spaces to achieve approximately the same expected time-discounted rewards. We first provide a decomposition theorem for return disparity, which decomposes the return disparity of any two MDPs sharing the same state and action spaces into the distance between group-wise reward functions, the discrepancy of group policies, and the discrepancy between state visitation distributions induced by the group policies. Motivated by our decomposition theorem, we propose algorithms to mitigate return disparity via learning a shared group policy with state visitation distributional alignment using integral probability metrics. We conduct experiments to corroborate our results, showing that the proposed algorithm can successfully close the disparity gap while maintaining the performance of policies on two real-world recommender system benchmark datasets.