Skip to content

This repository is a python implementation of ergodic control using tensor train format.

Notifications You must be signed in to change notification settings

SuhanNShetty/Ergodic_Exploration_using_Tensor_Train

Repository files navigation

Ergodic Exploration using Tensor Train (E2T2)

This repository is a python implementation of ergodic control using tensor train format corresponding to the paper:

  • "Ergodic Exploration using Tensor Train: Applications in Insertion Tasks" by Suhan Shetty, ‪João Silvério and Sylvain Calinon

Website: https://sites.google.com/view/ergodic-exploration/

Use the notebook Ergodic_Exploration_using_TT.ipynb. You can use google-colab to run this notebook (copy the url of this github repo in https://colab.research.google.com/notebooks). If you want further speed up (an order of magnitude speed-up in preprocessing step (i.e., computation of the Fourier coefficients) than that was reported in our paper), use the JAX (https://github.com/google/jax) powered notebook Ergodic_Exploration_using_TT_with_JAX.ipynb.

About

This repository is a python implementation of ergodic control using tensor train format.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published