A Novel Reinforcement Learning Architecture for Continuous State and Action Spaces

Figure 4

The dribbling problem.