Skip to content

Artificial curiosity algorithm using an approximate entropy lower bound based on k-means to explore continuous spaces in reinforcement learning.

Notifications You must be signed in to change notification settings

axnedergaard/kme

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 

Repository files navigation

k-Means Maximum Entropy Exploration

https://arxiv.org/abs/2205.15623

The directory mujoco contains code to reproduce exploration experiments and the directory kme contains

  • a C++ implementation of KME
  • a Python wrapped library for the C++ implementation
  • code to reproduce entropy experiments

To install

cd kme && make && pip install .

To run entropy experiments, use the compiled binary kme/entropy. To see options

./entropy --help  

To run exploration experiments, use the script mujoco/train.py. You must install the modified stable-baselines3 repository found in mujoco/libs. The script requires the environment variable DEVICE to set the PyTorch device. To see options

python train.py --help

Please refer to the paper for parameter values.

About

Artificial curiosity algorithm using an approximate entropy lower bound based on k-means to explore continuous spaces in reinforcement learning.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published