REINFORCE issues

In the previous section, we discussed the REINFORCE method, which is a natural extension of cross-entropy from Chapter 4, The Cross-Entropy Method. Unfortunately, both REINFORCE and cross-entropy still suffer from several problems, which make both of them limited to simple environments.

Full episodes are required

First of all, we still need to wait for the full episode to complete before we can start training. Even worse, both REINFORCE and cross-entropy behave better with more episodes used for training (just from the fact that more episodes mean more training data, which means more accurate PG). This situation is fine for short episodes in the CartPole, when in the beginning, we can barely handle the bar for more than 10 steps, but in Pong, it is completely different: every episode can lasts hundreds or even thousands of frames. It’s equally bad from the training perspective, as our training batch becomes very large and from sample efficiency, when we need to communicate...