search

RicardoDominguez / PyCREPS

Contextual Relative Entropy Policy Search for Reinforcement Learning in Python
15 stars 1 forks source link

Automatic calculation of dual function gradient #14

Open RicardoDominguez opened 6 years ago

RicardoDominguez commented 6 years ago

Using torch and theano.

RicardoDominguez commented 6 years ago

Automatic gradient computation for the dual function resulted in (as tested with 065113ae9678b1614bfb4c3b97d037c9e971c6b2):

Moving forward, check if this is also true with torch on the GPU.