wangdan269/pg_rnn

There are few caveats when you want to use a Recurrent Neural Network (RNN) policy with Policy Gradient Algorithms. This repository explains them and provide a solution for them. Please see the blog for more details.

Python