Optimal Skipping Rates: Training Agents with Fine-Grained Control Using Deep Reinforcement Learning

Figure 3

The proposed convolutional neural network (CNN) architecture employed in experiments.