Name		Name	Last commit message	Last commit date
gym_goal		gym_goal
gym_platform		gym_platform
gym_soccer		gym_soccer
.gitignore		.gitignore
README.md		README.md
config.py		config.py
evaluate.py		evaluate.py
plot.py		plot.py
plot_baselines.py		plot_baselines.py
run.py		run.py
stone.py		stone.py

Repository files navigation

CDRL

This repository includes several reinforcement learning algorithms for parameterised action space MDPs:

PA-DDPG [Hausknecht & Stone 2016]

All implementations of algorithms are based on Stable-baselines3 (https://github.com/DLR-RM/stable-baselines3). The plot.py borrows some code from Spinning Up (https://github.com/openai/spinningup)

Domains

Experiment scripts are provided to run each algorithm on the following domains with parameterised actions:

Platform (https://github.com/cycraig/gym-platform)
Robot Soccer Goal (https://github.com/cycraig/gym-goal)
Half Field Offense (https://github.com/cycraig/gym-soccer)

About

No description, website, or topics provided.

0 stars

1 watching

Report repository

Releases

Packages

Contributors

Languages