Cooperative multi-agent reinforcement learning for high-dimensional nonequilibrium control Official implementation of: Cooperative multi-agent reinfor