Open PeiYingjun opened 6 years ago
Well, I am not sure, but it seems that lots of actor-critic architecture shared the core layers and just use different heads. The two parts need to understand the environment, so sharing the features of the world might be faster for training. If you have time to try it, do not hesitate to make a PR
Exactly, I'm trying to rewrite the code
Thanks for reply, I have been busy at another project last few days, recently I get spare time. I have noticed that at comm_net, the variables of communication part(maybe along with encoder part) are not shared between critic and actor, I don't know whether it should be like these way in regular algorithms trained by DDPG like comm_net?