Reinforcement Learning with Probabilistic Boolean Network Models of Smart Grid Devices

Figure 4

Maximum expected reward for the RL agent interacting with the environment.