search

LARG / escape_room

An implementation of the Escape Room domain for Hierarchical Reinforcement Learning.
MIT License
24 stars 3 forks source link

Readme #2

Open p-christ opened 5 years ago

p-christ commented 5 years ago

Hi, I really loved your paper Escape Room: A Configurable Testbed for Hierarchical Reinforcement Learning and am looking to use your environments in some research I am doing.

I am a bit confused about exactly how to use the environments though and was wondering if you could possibly help me out? If I want to access the environments in a manner similar to how I would access an Open AI gym environment which file or function should I use?

I looked at the README to try and figure this out but it doesn't seem to have any information on this.

Thanks!

jmenashe commented 5 years ago

Hello @p-christ ,

Do you have an environment in mind already or do you want to use the button-press domain that we've already designed for the escape room? If the latter then I can provide you with a script that can run an experiment with a gym-compatible agent.

p-christ commented 5 years ago

thanks for getting back to me! Yep i just wanted to use the button-pressing games you described in your paper if possible?

p-christ commented 5 years ago

let me know if this is ok? I've almost finished building a hierarchical agent that I would love to try out on this environment