Deep Reinforcement Learning-Based Content Placement and Trajectory Design in Urban Cache-Enabled UAV Networks

Figure 3

Convergence of DDQN algorithm with different learning rates versus the number of training episodes.