aseembits93 / p1_navigation

Udacity Deep Reinforcement Nanodegree Assignment 1 Solution

Project 1: Navigation

Introduction

For this project, you will 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, your agent must get an average score of +13 over 100 consecutive episodes.

Getting Started

1. Download the 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

   (For Windows users) Check out this link if you need help with determining if your computer is running a 32-bit version or 64-bit version of the Windows operating system.

   (For AWS) If you'd like to train the agent on AWS (and have not enabled a virtual screen), then please use this link to obtain the environment.

2. Place the file in the DRLND GitHub repository, in the p1_navigation/ folder, and unzip (or decompress) the file.

3. Setup your Anaconda Environment as mentioned here

4. python train_banana.py

Solution

1. The vanilla DQN agent with default hyperparameters solves the environment in around 371 episodes.

Experiments done

1. Role of Neural Network Architecture on convergence - I tried different variants for the Q network - Increasing depth and decreasing width
2. Role of BatchNorm Layer - Did not help much with convergence.
3. Role of Residual Connection - Again, did not help much with convergence.

Conclusion

Simpler neural network architectures tend to have better convergence properties.

Experiments TODO

1. Role of Priority Experience Replay Buffer
2. Role of Double DQN
3. Role of Dueling DQN
4. Combining multiple improvements mentioned above.
5. Role of Rainbow DQN
6. Role of different Optimizers - eg. RMSProp, SGD
7. Explore 2nd order optimizers - BFGS-L