Train an agent to navigate (and collect bananas!) in a large, square world.
A reward of +1 is provided for collecting a yellow banana, and a reward of -1 is provided for collecting a blue banana. Thus, the goal of your agent is to collect as many yellow bananas as possible while avoiding blue bananas.
The state space has 37 dimensions and contains the agent's velocity, along with ray-based perception of objects around agent's forward direction. Given this information, the agent has to learn how to best select actions. Four discrete actions are available, corresponding to:
0
- move forward.1
- move backward.2
- turn left.3
- turn right.
The task is episodic, and in order to solve the environment, the agent must get an average score of +13 over 100 consecutive episodes.
-
Clone this repository. I am running on mac, so the unity runtime environment (Banana.app) is already in the repository
-
If you are running in Linux or windows, download the appropriate environment from one of the links below. You need only select the environment that matches your operating system:
- Linux: click here
- Mac OSX: click here
- Windows (32-bit): click here
- Windows (64-bit): click here
-
Place the file in the
DQN_Navigation/
folder, and unzip (or decompress) the file.
- One can run the iPython notebook & train the agent (slow) or
- use the iPython notebook to run a saved model (fast) or
- watch the video
DQN_Nav.m4v
(fastest)
The Report.pdf
has a summary of the algorithm, the implementation and the experimentation (network architecture, hyperparameter search et al)