Title: Rethinking Fairness for Human-AI Collaboration
Abstract: Existing approaches to algorithmic fairness aim to ensure equitable outcomes if human decision-makers comply perfectly with algorithmic decisions. However, perfect compliance with the algorithm is rarely a reality or even a desirable outcome in human-AI collaboration. Yet, recent studies have shown that selective compliance with fair algorithms can amplify discrimination relative to the prior human policy. As a consequence, ensuring equitable outcomes requires fundamentally different algorithmic design principles that ensure robustness to the decision-maker's (a priori unknown) compliance pattern. We define the notion of compliance-robustly fair algorithmic recommendations that are guaranteed to (weakly) improve fairness in decisions, regardless of the human's compliance pattern. We propose a simple optimization strategy to identify the best performance-improving compliance-robustly fair policy. However, we show that it may be infeasible to design algorithmic recommendations that are simultaneously fair in isolation, compliance-robustly fair, and more accurate than the human policy; thus, if our goal is to improve the equity and accuracy of human-AI collaboration, it may not be desirable to enforce traditional fairness constraints. [Joint work with H. Ge and O. Bastani; extended abstract to appear in ITCS 2024. Paper: https://arxiv.org/abs/2310.03647]
Bio: Hamsa Bastani is an Associate Professor of Operations, Information and Decisions as well as Statistics and Data Science at the Wharton School of the University of Pennsylvania, where she co-directs the Wharton Healthcare Analytics Lab. Her research primarily focuses on developing novel machine learning algorithms for dynamic learning and optimization for social good. She has worked closely with national governments, including Greece and Sierra Leone, to deploy algorithms at the country-scale to improve public health outcomes. Her research has been published in leading outlets including Nature, Management Science, and Operations Research, and has received several recognitions, including the Wagner Prize, the Pierskalla Award, and the George Nicholson Prize.
