Fairer AI evaluation via game theory
A new framework uses game theory to evaluate AI agents more fairly when human preferences differ across contexts.
A new framework uses game theory to evaluate AI agents more fairly when human preferences differ across contexts.