Adjusting Machine Learning Decisions for Equal Opportunity and Counterfactual Fairness

Yixin Wang, Dhanya Sridhar, David Blei

Abstract:

Machine learning (ML) methods have the potential to automate
high-stakes decisions, such as bail admissions or credit lending, by
analyzing and learning from historical data. But these algorithmic
decisions may be unfair: in learning from historical data, they may
replicate discriminatory practices from the past. In this paper, we
propose two algorithms that adjust fitted ML predictors to produce
decisions that are fair. Our methods provide post-hoc adjustments to
the predictors, without requiring that they be retrained. We consider
a causal model of the ML decisions, define fairness through
counterfactual decisions within the model, and then form algorithmic
decisions that capture the historical data as well as possible but
are provably fair. In particular, we consider two definitions of
fairness. The first is ``equal counterfactual opportunity,'' where
the counterfactual distribution of the decision is the same regardless
of the protected attribute; the second is counterfactual fairness. We
evaluate the algorithms, and the trade-off between accuracy and
fairness, on datasets about admissions, income, credit, and
recidivism.

Submission Length: Regular submission (no more than 12 pages of main content)

Changes Since Last Submission: Address all meta reviewer comments: Clarify the optimality of FTU in the discussion; clarify the difference between FTU and ECO early on in the discussion; clarify the choice of the population considered in the expectation (intended as a constraint); change notations per meta reviewer's suggestions; address all other comments

Assigned Action Editor: ~Alexandra_Chouldechova1

License: Creative Commons Attribution 4.0 International (CC BY 4.0)

Submission Number: 445