[Question]: Reference for the Policy Gradient implementation

Hi Andrej,

first of all thank you for sharing with us your code and for your really interesting blog that contains a lot of valuable resources for students like me :)

I've carefully studied the code reported in this repository in order to better understand how to implement Policy Gradient. It was a bit hard to follow the original *REINFORCE* paper by Williams but your implementation and the one reported in the [*rllab* library](https://github.com/openai/rllab), give me the possibility to implement the algorithm.

I have seen that you use a lot of tricks in order to stabilize and enhance the original version of the algorithm. Could you please share with us a reference for the implementation that you have written? I think that can be really useful in order to fully understand the code and the reasons behind the tricks that you've applied.

Thank you for your availability.

Best regards,
Alessandro