The Blog



Share

Why do Policy Gradient Methods work so well in Cooperative MARL? Evidence from Policy Representation