Skip to content

Code for the paper: Graph-Based Design of Hierarchical Reinforcement Learning Agents

Notifications You must be signed in to change notification settings

AIRLab-POLIMI/mushroom_hierarchical

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Mushroom Hierarchical

Mushroom Hierarchical: Hierarchical Reinforcement Learning python library.

This repository implements the hierarchical framework proposed in the paper: "A Control Theoretic Approach to Hierarchical Reinforcement Learning"

What is Mushroom Hierarchical

Mushroom Hierarchical is a python Hierarchical Reinforcement Learning (HRL) library based on Mushroom, the RL library of Politecnico di Milano. It allows to perform HRL experiments exploiting the control graph formalism.

Installation

You can do a minimal installation of Mushroom with:

git clone https://github.com/AIRLab-POLIMI/mushroom.git
cd mushroom
pip3 install -e .

You can install Mushroom Hierarchical with:

git clone https://github.com/AIRLab-POLIMI/mushroom_hierarchical.git
cd mushroom_hierarchical
pip3 install -e .

How to set and run and experiment

To run experiments, you should use the run.py script in the experiment folder. If you want to visualize the results produced by our runs, you can look at the out folder, or you can re-run the plot_*.py scripts into the graphs folder.

About

Code for the paper: Graph-Based Design of Hierarchical Reinforcement Learning Agents

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published