# Fair Policy Targeting

arXiv: https://arxiv.org/abs/2005.12395 · HTML: https://mlsanity.com/paper/2005.12395/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 25 May 2020.

## Built on by (most cited first, 12 of 12)

- [Learning to be Fair: A Consequentialist Approach to Equitable Decision-Making](https://mlsanity.com/paper/2109.08792/) (2021)
- [Fair Off-Policy Learning from Observational Data](https://mlsanity.com/paper/2303.08516/) (2023)
- [When Personalization Harms: Reconsidering the Use of Group Attributes in Prediction](https://mlsanity.com/paper/2206.02058/) (2022)
- [Fair and Robust Estimation of Heterogeneous Treatment Effects for Policy Learning](https://mlsanity.com/paper/2306.03625/) (2023)
- [Participatory Personalization in Classification](https://mlsanity.com/paper/2302.03874/) (2023)
- [Counterfactually Fair Reinforcement Learning via Sequential Data Preprocessing](https://mlsanity.com/paper/2501.06366/) (2025)
- [Econometrics of Machine Learning Methods in Economic Forecasting](https://mlsanity.com/paper/2308.10993/) (2023)
- [Towards Optimal Environmental Policies: Policy Learning under Arbitrary Bipartite Network Interference](https://mlsanity.com/paper/2410.08362/) (2024)
- [Fairness-aware kidney exchange and kidney paired donation](https://mlsanity.com/paper/2503.06431/) (2025)
- [Optimal Transport Learning: Balancing Value Optimization and Fairness in Individualized Treatment Rules](https://mlsanity.com/paper/2507.23349/) (2025)
- [Dynamic Treatment on Networks](https://mlsanity.com/paper/2605.06564/) (2026)
- [Counterfactually Safe Reinforcement Learning](https://mlsanity.com/paper/2605.25114/) (2026)

## Abstract (arXiv)

One of the major concerns of targeting interventions on individuals in social welfare programs is discrimination: individualized treatments may induce disparities across sensitive attributes such as age, gender, or race. This paper addresses the question of the design of fair and efficient treatment allocation rules. We adopt the non-maleficence perspective of first do no harm: we select the fairest allocation within the Pareto frontier. We cast the optimization into a mixed-integer linear program formulation, which can be solved using off-the-shelf algorithms. We derive regret bounds on the unfairness of the estimated policy function and small sample guarantees on the Pareto frontier under general notions of fairness. Finally, we illustrate our method using an application from education economics.