Hands-On Reinforcement Learning for Games

Exercises

The exercises in this section are for you to explore on your own. Substantially advancing any of the techniques we cover from this point forward is an accomplishment, so the work you do here could morph into something beyond just learning. Indeed, the environments and examples you work on now will likely indicate your working preference going forward. As always, try to complete two to three of the following exercises:

1. Tune the hyperparameters for Chapter_9_PPO.py and/or Chapter_9_PPO_LSTM.py.
2. Tune the hyperparameters for Chapter_9_A2C.py and/or Chapter_9_A3C.py.
3. Tune the hyperparameters for Chapter_9_ACER.py.
4. Apply LSTM layers to the A2C and/or A3C examples.
5. Apply LSTM layers to the ACER example.
6. Add a play_game function to the ...