The Sensitivity of Counterfactual Fairness to Unmeasured Confounding

By Niki Kilbertus, Philip J. Ball, Matt J. Kusner, Adrian Weller, Ricardo M. A. Silva. 2019

65 citations recorded by the source corpus.

Causal approaches to fairness have seen substantial recent interest, both from the machine learning community and from wider parties interested in ethical prediction algorithms. In no small part, this has been due to the fact that causal models allow one to simultaneously leverage data and expert knowledge to remove discriminatory effects from predictions. However, one of the primary assumptions in causal modeling is that you know the causal graph. This introduces a new opportunity for bias, caused by misspecifying the causal model. One common way for misspecification to occur is via unmeasured confounding: the true causal effect between variables is partially described by unobserved quantities. In this work we design tools to assess the sensitivity of fairness measures to this confounding for the popular class of non-linear additive noise models (ANMs). Specifically, we give a procedure for computing the maximum difference between two counterfactually fair predictors, where one has become biased due to confounding. For the case of bivariate confounding our technique can be swiftly computed via a sequence of closed-form updates. For multivariate confounding we give an algorithm that can be efficiently solved via automatic differentiation. We demonstrate our new sensitivity analysis tools in real-world fairness scenarios to assess the bias arising from confounding.

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): “Of particular interest to our work is the notion of counterfactual fairness [9] (and extensions [10, 16–18]), which calls a (probabilistic) classifier h over V = X ∪A counterfactually fair if it satisfies h(vF) = h(va(u F)) for all a ∈ A and vF = (xF, aF) ∈ X ×A, where va(u) denotes the “counterfact”
Counterfactual Fairness Is Basically Demographic Parity (2022, background): “In (Kilbertus et al. 2020), they examine the sensitivity of counterfactual fairness to “un-measured confounding.””