Double Fairness Policy Learning: Integrating Action Fairness and Outcome Fairness in Decision-making

TL;DR

This paper introduces a double fairness learning framework that balances action fairness, outcome fairness, and value maximization in policy learning, with theoretical guarantees and practical improvements demonstrated through simulations and real datasets.

Contribution

It presents a novel multi-objective optimization approach for policy learning that explicitly manages two types of fairness and provides theoretical and empirical validation.

Findings

Improved action and outcome fairness in simulations and real datasets.

The framework effectively balances fairness and value with minimal performance loss.

The method offers theoretical guarantees on regret bounds.

Abstract

Fairness is a central pillar of trustworthy machine learning, especially in domains where accuracy- or profit-driven optimization is insufficient. While most fairness research focuses on supervised learning, fairness in policy learning remains less explored. Because policy learning is interventional, it induces two distinct fairness targets: action fairness (equitable action assignments) and outcome fairness (equitable downstream consequences). Crucially, equalizing actions does not generally equalize outcomes when groups face different constraints or respond differently to the same action. We propose a novel double fairness learning (DFL) framework that explicitly manages the trade-off among three objectives: action fairness, outcome fairness, and value maximization. We integrate fairness directly into a multi-objective optimization problem for policy learning and employ a lexicographic weighted Tchebyshev method that recovers Pareto solutions beyond convex settings, with theoretical guarantees on the regret bounds. Our framework is flexible and accommodates various commonly used fairness notions. Extensive simulations demonstrate improved performance relative to competing methods. In applications to a motor third-party liability insurance dataset and an entrepreneurship training dataset, DFL substantially improves both action and outcome fairness while incurring only a modest reduction in overall value.