RLearning :
Exploit Reinforcement Learning using OpenAI api
Description :
I implemented Q learning using simple q-table (with Bellman equation)
And Try to define deep q learning with a neural net
TODO:
- implemented a simple Deep Q Learning (Keras)
- Try to generalize the pipeline (any use case or large use case)
Usage :
python3 tests/test_dqn.py --game Alien-ram-v0