This paper proposes a new architecture for multiagent systems to cover an unknowingly distributed fast, safely, and decentralizedly. The inter-agent communication is organized by a directed graph with fixed topology, and we model agent coordination as a decentralized leader-follower problem with time-varying communication weights. Given this problem setting, we first present a method for converting communication graph into a neural network, where an agent can be represented by a unique node of the communication graph but multiple neurons of the corresponding neural network. We then apply a mass-cetric strategy to train time-varying communication weights of the neural network in a decentralized fashion which in turn implies that the observation zone of every follower agent is independently assigned by the follower based on positions of in-neighbors. By training the neural network, we can ensure safe and decentralized multi-agent coordination of coverage control. Despite the target is unknown to the agent team, we provide a proof for convergence of the proposed multi-agent coverage method. The functionality of the proposed method will be validated by a large-scale multicopter team covering distributed targets on the ground.