Continuous Time Natural Actor Critic Reinforcement Learning
This repository is a basic actor-critic reinforcement learning implementation based on the following papers:
http://www.ualberta.ca/~pilarski/docs/papers/Pilarski_2011_ICORR.pdf
and
http://www.ualberta.ca/~pilarski/docs/papers/Pilarski_2013_ICORR_Postprint.pdf
How can ACRL be used for multi-joint control of robotic limbs?