Supplements of " Non-crossing quantile regression in deep reinforcement learning "