Exercises

These exercises are here for you to use and learn from. Attempt at least 2-3, and the more you do, the easier later chapters will also be:

1. What is the difference between an online and offline policy agent?
2. Tune the hyperparameters for any or all of the examples in this chapter, including the new hyperparameter, lambda.
3. Change the discretization steps in any example that uses discretization and see what effect it has on training.
4. Use example Chapter_5_3.py, SARSA(0), and adapt it to another Gym environment that uses a continuous observation space and discrete action space.
5. Use example Chapter_5_4.py, SARSA(λ), and adapt it to another Gym environment that uses a continuous observation space and discrete action space.
6. There is a hyperparameter shown in the code that is not used. Which parameter is it?
7. Use example Chapter_5_5.py, SARSA(λ), Lunar Lander and optimize...