Security and Communication Networks

Research Article

Early Rumor Detection Based on Deep Recurrent Q-Learning

Figure 6

Reward value changing state during training.