Solution to Exercise 2.1

The difference is quite substantial: with a trained value function, the agent is able to quickly make progress. With an untrained value function, the agent gets stuck early on.

← 返回课程

spinningup-zh

21 / 30·markdown

Chapter 21

Solution to Exercise 2.1

The difference is quite substantial: with a trained value function, the agent is able to quickly make progress. With an untrained value function, the agent gets stuck early on.