9 More stable value-based methods

 

In this chapter

  • You will improve on the methods you learned in the previous chapter by making them more stable and therefore less prone to divergence.
  • You will explore advanced value-based deep reinforcement learning methods, and the many components that make value-based methods better.
  • You will solve the cart-pole environment in a fewer number of samples, and with more reliable and consistent results.

Let thy step be slow and steady, that thou stumble not.

— Tokugawa Ieyasu Founder and first shōgun of the Tokugawa shogunate of Japan and one of the three unifiers of Japan

DQN: Making reinforcement learning more like supervised learning

Common problems in value-based deep reinforcement learning

Using target networks

Using larger networks

Using experience replay

Using other exploration strategies

Double DQN: Mitigating the overestimation of action-value functions

The problem of overestimation, take two

Separating action selection from action evaluation

A solution

A more practical solution

A more forgiving loss function

Things we can still improve on

Summary

sitemap