Interventional Fairness: Causal Database Repair for Algorithmic Fairness

By Babak Salimi, Luke Rodriguez, Bill Howe, Dan Suciu. 2019

213 citations recorded by the source corpus.

Fairness is increasingly recognized as a critical component of machine learning systems. However, it is the underlying data on which these systems are trained that often reflect discrimination, suggesting a database repair problem. Existing treatments of fairness rely on statistical correlations that can be fooled by statistical anomalies, such as Simpson's paradox. Proposals for causality-based definitions of fairness can correctly model some of these situations, but they require specification of the underlying causal models. In this paper, we formalize the situation as a database repair problem, proving sufficient conditions for fair classifiers in terms of admissible variables as opposed to a complete causal model. We show that these conditions correctly capture subtle fairness violations. We then use these conditions as the basis for database repair algorithms that provide provable fairness guarantees about classifiers trained on their training labels. We evaluate our algorithms on real data, demonstrating improvement over the state of the art on multiple fairness metrics proposed in the literature while retaining high utility.

Subfields: Artificial Intelligence, Safety Research

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Safety Research

Citations in the corpus

On the Fairness of Causal Algorithmic Recourse (2020, background): “2017a,b), these are sometimes mutually incompatible (Chouldechova 2017), and it has been argued that discrimination, at its heart, corresponds to a (direct or indirect) causal influence of a protected attribute on the prediction, thus making fairness a fundamentally causal problem (Kilbertus et al. ”
An Intersectional Definition of Fairness (2018, background): “The AI community has responded by developing a broad array of mathematical formulations of fairness and learning algorithms which aim to satisfy them [3], [10], [13], [20], [24].”
A Review on Fairness in Machine Learning (2022, background): “[178, 179] suggest to use causal knowledge on dependencies between sensitive and non-sensitive attributes to add or remove samples from the training dataset in a manner that improves fairness.”