XY Universe: A 2D Particle Survival Environment for Deep Reinforcement Learning
source link: https://www.tuicool.com/articles/hit/7FN773Y
Go to the source link to view the article. You can view the picture content, updated content and better typesetting reading experience. If the link is broken, please click the button below to view the snapshot at that time.
We provide here XY Universe: a 2D particle survival environment for training your Deep Reinforcement Learning agents to stay alive as long as possible by avoiding collisions with “bad” particles. The agent and the “bad” particles are confined to a 2D box, move with fixed constant speed, and experience elastic collisions against the box walls. A Deep Q-Network agent implementation is provided for immediate usage and demonstration of the environment. The agent reward scheme is: +1 for each time step alive, and -100 for “bad” particle collision. The episode termination condition is: agent collision with any “bad” particle. This survival environment was inspired by Karpathy’s Waterworld environment .
The XY Universe survival particle environment and DQN implementation can be found here:
https://github.com/ankonzoid/XY_universe
For more of my blogs, tutorials, and projects on Machine Learning, Deep Learning and Reinforcement Learning, please check my Medium and my Github .
Recommend
About Joyk
Aggregate valuable and interesting links.
Joyk means Joy of geeK