Cooperative Multi-Agent Fairness and Equivariant Policies

TL;DR

This paper introduces a novel approach to fairness in cooperative multi-agent learning by enforcing equivariant policies, demonstrating improved fairness and utility trade-offs through empirical validation.

Contribution

It proposes the concept of team fairness, develops the Fairness through Equivariance (Fair-E) strategy, and introduces a regularized version Fair-ER, advancing fairness in multi-agent systems.

Findings

Fair-ER achieves higher utility than Fair-E.

Fair policies result in fairer outcomes than non-equivariant policies.

The fairness-utility trade-off depends on agent skill.

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 prove that it is possible to enforce team fairness during policy optimization by transforming the team's joint policy into an equivariant map. We refer to our multi-agent learning strategy as Fairness through Equivariance (Fair-E) and demonstrate its effectiveness empirically. We then introduce Fairness through Equivariance Regularization (Fair-ER) as a soft-constraint version of Fair-E and show that it reaches higher levels of utility than Fair-E and fairer outcomes than non-equivariant policies. Finally, we present novel findings regarding the fairness-utility trade-off in multi-agent settings; showing that the magnitude of the trade-off is dependent on agent skill.