In this project, we 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. The goal of our 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 is considered solved when our agent is able to get an average score of +13 over 100 consecutive episodes.
-
Run
python setup.pyto install the dependencies -
Download the unity ml environment from one of the links below and extract it under env folder. 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
The repository is organised as following:
env- stores all the unity ml environmentsnotebooks- stores the notebooks for environment exploration, training agents and hyperparameter tuning.saved_models- stores saved network weightssrc- includes the code for agent and neural networks. agents.py contains all the agents, model.py contains neural network implementation and player.py allows us to run the trained model.
👉 Use notebooks/Trainer.ipynb to see training procedure, experiments with hyperparameters.
👉 Report.ipynb includes the information of how the agent was trained, its performance against a player that chooses actions randomly and future ideas.
Note: The training and testing was performed on local Ubuntu 16.04.
✅ DQN Agent
◻️ Double DQN Agent
◻️ Agent with Prioritized Experience Replay
◻️ Dueling DQN Agent