in these repo we use DQN to train an agent within unity env to learn how to navigate and get the right banana color
the DQN is a Deep Neural network that get as input the env state and give as an action we use the max to choice the the max action value
he project environment is similar to, but not identical to the Banana Collector environment on the Unity ML-Agents GitHub page.
farther impovment can be done by impleent double DQN
for ore inofrmation about DQN please read these paper https://storage.googleapis.com/deepmind-media/dqn/DQNNaturePaper.pdf