Skip to content
A TensorFlow implementation of https://arxiv.org/abs/1509.08731 by S. Mohamed and D. Rezende
Branch: master
Clone or download
Latest commit db18a5f Dec 24, 2018
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
__pycache__ add net action Dec 23, 2018
checkpoints add net action Dec 23, 2018
experiments add net action Dec 23, 2018
heat_maps update Dec 23, 2018
tests
.DS_Store add net action Dec 23, 2018
README.md
__init__.py first commit Nov 25, 2018
agent.py add net action Dec 23, 2018
main.py update Dec 23, 2018
setup.py add ildefons Nov 25, 2018
square_env.py nothing works Dec 2, 2018
utils.py add gradient norm expt Nov 29, 2018
visualisation.py nothing works Dec 2, 2018

README.md

Motivation:

  1. To implement the paper of Shakir Mohamed and Danilo Rezende https://arxiv.org/abs/1509.08731 aka 'Variational Information Maximisation for Intrinsically Motivated Reinforcement Learning'
  2. To develop an understanding of the value and limitations of Empowerment for Intrinsic Control.
  3. Empowerment was originally introduced by Salge, Glackin and Polani here: https://arxiv.org/abs/1310.1863

Note: This project was initially co-developed with Ildefons Magrans(Ildefons.magrans@gmail.com).

You can’t perform that action at this time.