Algorithmic systems, designed to streamline decision processes and enhance efficiency, have permeated virtually every aspect of our lives. From credit approvals to hiring decisions, from predictive policing to healthcare recommendations, algorithms wield significant influence. Yet, this influence is not neutral, and the consequences could be disproportionate for diverse communities. Subtle biases embedded in training data, the choices made during model development, and the very nature of algorithmic decision-making are some potential reasons for inequitable treatment of certain demographic groups, perpetuating and, in some instances, exacerbating societal disparities. Consider, for instance, the use of predictive policing algorithms, where certain communities are subjected to heightened surveillance based on historical crime data, perpetuating a cycle of over-policing [9]. Similarly, in hiring practices, algorithms may inadvertently favor certain demographics, leading to underrepresentation and reinforcing existing inequalities in the workplace [6, 5]. Therefore, it is crucial to acknowledge the inherent biases and disparities that have emerged within these systems and propose innovative solutions to enhance their fairness guarantees.

The importance of incorporating ethics and legal compliance into machine-assisted decision-making is broadly recognized. Further, several lines of recent work have argued that critical opportunities for improving data quality and representativeness, controlling for bias, and allowing humans to oversee and impact computational processes are missed if we do not consider the lifecycle stages upstream from model training and deployment. Yet, very little has been done to date to provide system-level support to data scientists who wish to develop and deploy responsible machine learning methods. We aim to fill this gap and present FairPrep, a design and evaluation framework for fairness-enhancing interventions. FairPrep is based on a developer-centered design, and helps data scientists follow best practices in software engineering and machine learning. As part of our contribution, we identify shortcomings in existing empirical studies for analyzing fairness-enhancing interventions. We then show how FairPrep can be used to measure the impact of sound best practices, such as hyperparameter tuning and feature scaling. In particular, our results suggest that the high variability of the outcomes of fairness-enhancing interventions observed in previous studies is often an artifact of a lack of hyperparameter tuning. Further, we show that the choice of a data cleaning method can impact the effectiveness of fairness-enhancing interventions.

The wide spread usage of automated data-driven decision support systems has raised a lot of concerns regarding accountability and fairness of the employed models in the absence of human supervision. Existing fairness-aware approaches tackle fairness as a batch learning problem and aim at learning a fair model which can then be applied to future instances of the problem. In many applications, however, the data comes sequentially and its characteristics might evolve with time. In such a setting, it is counter-intuitive to "fix" a (fair) model over the data stream as changes in the data might incur changes in the underlying model therefore, affecting its fairness. In this work, we propose fairness-enhancing interventions that modify the input data so that the outcome of any stream classifier applied to that data will be fair. Experiments on real and synthetic data show that our approach achieves good predictive performance and low discrimination scores over the course of the stream.

Computers are increasingly used to make decisions that have significant impact in people's lives. Often, these predictions can affect different population subgroups disproportionately. As a result, the issue of fairness has received much recent interest, and a number of fairness-enhanced classifiers and predictors have appeared in the literature. This paper seeks to study the following questions: how do these different techniques fundamentally compare to one another, and what accounts for the differences? Specifically, we seek to bring attention to many under-appreciated aspects of such fairness-enhancing interventions. Concretely, we present the results of an open benchmark we have developed that lets us compare a number of different algorithms under a variety of fairness measures, and a large number of existing datasets. We find that although different algorithms tend to prefer specific formulations of fairness preservations, many of these measures strongly correlate with one another. In addition, we find that fairness-preserving algorithms tend to be sensitive to fluctuations in dataset composition (simulated in our benchmark by varying training-test splits), indicating that fairness interventions might be more brittle than previously thought.