• Corpus ID: 210921184

Algorithmic Fairness

@article{Pessach2020AlgorithmicF,
  title={Algorithmic Fairness},
  author={Dana Pessach and Erez Shmueli},
  journal={ArXiv},
  year={2020},
  volume={abs/2001.09784}
}
An increasing number of decisions regarding the daily lives of human beings are being controlled by artificial intelligence (AI) algorithms in spheres ranging from healthcare, transportation, and education to college admissions, recruitment, provision of loans and many more realms. Since they now touch on many aspects of our lives, it is crucial to develop AI algorithms that are not only accurate but also objective and fair. Recent studies have shown that algorithmic decision-making may be… 

Figures and Tables from this paper

CONFAIR: Configurable and Interpretable Algorithmic Fairness

The CONFAIR procedure is proposed that produces a fair algorithm by incorporating user constraints into the optimization procedure and is demonstrated the efficacy of the approach on several real world datasets using different fairness criteria.

FAIRLEARN: Configurable and Interpretable Algorithmic Fairness

The FAIRLEARN procedure is proposed that produces a fair algorithm by incorporating user constraints into the optimization procedure and is demonstrated the efficacy of the approach on several real world datasets using different fairness criteria.

Coping with Mistreatment in Fair Algorithms

This paper studies the algorithmic fairness in a supervised learning setting and examines the effect of optimizing a classifier for the Equal Opportunity metric and proposes a conceptually simple method to mitigate this bias.

A Statistical Test for Probabilistic Fairness

This paper develops a rigorous hypothesis testing mechanism for assessing the probabilistic fairness of any pre-trained logistic classifier, and shows both theoretically as well as empirically that the proposed test is asymptotically correct.

Algorithmic Fairness in a Technology-Based World

This literature review seeks to break the field into its sub-components and delve into the research being done in each and convey the real-world impacts decision-making algorithms are increasingly involved in across a range of domains.

On the Advantages of Distinguishing Between Predictive and Allocative Fairness in Algorithmic Decision-Making

The problem of algorithmic fairness is typically framed as the problem of finding a unique formal criterion that guarantees that a given algorithmic decision-making procedure is morally permissible, but it is argued that this is conceptually misguided and that the problem should be replaced with two sub-problems.

Algorithm Design: Fairness and Accuracy

This work defines and characterize a fairness-accuracy frontier, consisting of the optimal points across a broad range of criteria for trading off fairness and accuracy, and identifies how the algorithm’s inputs govern the shape of this frontier.

Algorithmic Design: Fairness Versus Accuracy

A model in which a designer chooses an algorithm that maps observed inputs into decisions is proposed, and a fairness-accuracy Pareto frontier is introduced, showing (for example) that access to group identity reduces the error for the worse-off group everywhere along the frontier.

An Exploratory Study on Fairness-Aware Design Decision-Making

This paper quantifies unfairness and analyzes its impact in the context of data-driven engineering design using the Adult Income dataset, and introduces a fairness-aware design concept and standard definitions and statistical measures of fairness to the engineering design research.

Matching code and law: achieving algorithmic fairness with optimal transport

The explicit formalization of the trade-off between individual and group fairness allows this post-processing approach to be tailored to different situational contexts in which one or the other fairness criterion may take precedence.
...

References

SHOWING 1-10 OF 140 REFERENCES

A comparative study of fairness-enhancing interventions in machine learning

It is found that fairness-preserving algorithms tend to be sensitive to fluctuations in dataset composition and to different forms of preprocessing, indicating that fairness interventions might be more brittle than previously thought.

Algorithmic Decision Making and the Cost of Fairness

This work reformulate algorithmic fairness as constrained optimization: the objective is to maximize public safety while satisfying formal fairness constraints designed to reduce racial disparities, and also to human decision makers carrying out structured decision rules.

Mathematical Notions vs. Human Perception of Fairness: A Descriptive Approach to Fairness for Machine Learning

The most simplistic mathematical definition of fairness is found, demographic parity, which most closely matches people's idea of fairness in two distinct application scenarios.

Fairness in Decision-Making - The Causal Explanation Formula

The causal explanation formula is derived, which allows the AI designer to quantitatively evaluate fairness and explain the total observed disparity of decisions through different discriminatory mechanisms, and provides a quantitative approach to policy implementation and the design of fair AI systems.

Enhancing the Accuracy and Fairness of Human Decision Making

This paper addresses the above problem from the perspective of sequential decision making and shows that it reduces to a sequence of (constrained) weighted bipartite matchings, which can be solved efficiently using algorithms with approximation guarantees.

Fairness Under Composition

This work identifies pitfalls of naive composition and gives general constructions for fair composition, demonstrating both that classifiers that are fair in isolation do not necessarily compose into fair systems and also that seemingly unfair components may be carefully combined to construct fair systems.

Fairness Constraints: Mechanisms for Fair Classification

This paper introduces a flexible mechanism to design fair classifiers by leveraging a novel intuitive measure of decision boundary (un)fairness, and shows on real-world data that this mechanism allows for a fine-grained control on the degree of fairness, often at a small cost in terms of accuracy.

Active Fairness in Algorithmic Decision Making

active approaches can substantially outperform randomization-based classifiers previously considered optimal, while avoiding limitations such as intra-group unfairness.

Assessing algorithmic fairness with unobserved protected class using data combination

This paper studies a fundamental challenge to assessing disparate impacts, or performance disparities in general, in practice: protected class membership is often not observed in the data, particularly in lending and healthcare, and provides optimization-based algorithms for computing and visualizing sets of simultaneously achievable pairwise disparties for assessing disparities.

The Measure and Mismeasure of Fairness: A Critical Review of Fair Machine Learning

It is argued that it is often preferable to treat similarly risky people similarly, based on the most statistically accurate estimates of risk that one can produce, rather than requiring that algorithms satisfy popular mathematical formalizations of fairness.
...