Thank you for this awesome series.
Gabriel de la Cruz

Hi Gabriel,

Is the issue that the reward is no longer increasing, or that it is taking increasingly longer to complete episodes. In breakout one reason for this might be that as the episode length increases, the frequency at which results are posted to the tensorboard decreases, since it is tied to the episode count.

Like what you read? Give Arthur Juliani a round of applause.

From a quick cheer to a standing ovation, clap to show how much you enjoyed this story.