Evaluating and Rewarding Teamwork Using Cooperative Game Abstractions
This addresses the challenge of evaluating and incentivizing teamwork in domains like AI and sports, though it is incremental by building on cooperative game theory with computational improvements.
The paper tackles the problem of predicting team performance and fairly rewarding individual contributions by introducing cooperative game abstractions (CGAs), a parametric model that learns characteristic functions from data and enables linear-time computation of the Shapley Value, with applications showing effectiveness in RL agent teams and professional sports.
Can we predict how well a team of individuals will perform together? How should individuals be rewarded for their contributions to the team performance? Cooperative game theory gives us a powerful set of tools for answering these questions: the Characteristic Function (CF) and solution concepts like the Shapley Value (SV). There are two major difficulties in applying these techniques to real world problems: first, the CF is rarely given to us and needs to be learned from data. Second, the SV is combinatorial in nature. We introduce a parametric model called cooperative game abstractions (CGAs) for estimating CFs from data. CGAs are easy to learn, readily interpretable, and crucially allow linear-time computation of the SV. We provide identification results and sample complexity bounds for CGA models as well as error bounds in the estimation of the SV using CGAs. We apply our methods to study teams of artificial RL agents as well as real world teams from professional sports.