Fairness for Cooperative Multi-Agent Learning with Equivariant Policies.
Year: 2021
Type: preprint
Abstract: We study fairness through the lens of cooperative multi-agent learning. Our work is motivated by empirical evidence that naive maximization of team reward yields unfair outcomes for individual team members. To address fairness in multi-agent contexts, we introduce team fairness, a group-based fairness measure for multi-agent learning. We then incorporate team fairness into policy optimization -- introducing Fairness through Equivariance (Fair-E), a novel learning strategy that achieves provably fair reward distributions. We then introduce Fairness through Equivariance Regularization (Fair-ER) as a soft-constraint version of Fair-E and show that Fair-ER reaches higher levels of utility than Fair-E and fairer outcomes than policies with no equivariance. Finally, we investigate the fairness-utility trade-off in multi-agent settings.
Source: arXiv (Cornell University)
Authors: Niko A. Grupen, Bart Selman, Daniel D. Lee
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
SDG: Reduced inequalities
Open Access status: green
OpenAlex ID: W3168240225