A Dynamic Adjusting Reward Function Method for Deep Reinforcement Learning with Adjustable Parameters

Figure 1

Network structures of DQN and dueling DQN.