A2064
Title: Optimal transport learning: Balancing value optimization and fairness in individualized treatment rules
Authors: Wenhai Cui - The Hong Kong Polytechnic University (China)
Xiaoting Ji - The Hong Kong Polytechnic University (China) [presenting]
Wen Su - City University of Hong Kong (Hong Kong)
Xiaodong Yan - Xi An Jiaotong University (China)
Xingqiu Zhao - The Hong Kong Polytechnic University (Hong Kong)
Abstract: Individualized treatment rules (ITRs) have gained significant attention due to their wide-ranging applications in fields such as precision medicine, ridesharing, and social welfare distribution. However, when ITRs are influenced by sensitive attributes such as race, gender, or age, they can lead to outcomes where certain groups are unfairly advantaged or disadvantaged. To address this gap, a flexible approach based on optimal transport theory is proposed, which is capable of transforming any optimal ITR into a fair ITR that ensures demographic parity. A trade-off ITR is introduced, designed to balance value optimization and fairness while accommodating varying levels of fairness through parameter adjustment. Additionally, a theoretical upper bound on the value loss for the trade-off ITR is established. The proposed method is demonstrated through extensive simulation studies and application to the Oregon Health Insurance Experiment dataset.
