Share weights between first layer the actor network and first layer of the critic network for DDPG

[SharhadBashar]

Train an agent on the pendulum task

I want to Modify the DDPG code in baselines/ddpg/ to share the weights between the first layer of the actor network and the first layer of the critic network

Then I want to produce two graphs ofa verage reward of each episode obtained by DDPG in the Pendulum-v0 environment where I compare the unmodified DDPG algorithm, and the modified DDPG algorithm

How do I do that?