It seems that you don't use "Policy ensembles" and "Inferring policies of other agent" in this code?

openai / maddpg

Code for the MADDPG algorithm from the paper "Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environments"

https://arxiv.org/pdf/1706.02275.pdf

MIT License

1.66k stars 494 forks source link

It seems that you don't use "Policy ensembles" and "Inferring policies of other agent" in this code? #8

Closed pengzhenghao closed 6 years ago

pengzhenghao commented 6 years ago

I can't find the relative code here. So maybe I have left out them by my carelessness. So could you point it out?

ryan-lowe commented 6 years ago

Hi, You are correct -- this repo only contains the code for the core maddpg algorithm. The code for the policy ensembles and inferring other agent's policies was mostly done by Yi Wu --- if you are interested in that code, I'd recommend e-mailing him (jxwuyi@gmail.com).

jxwuyi commented 5 years ago

For easy access, I have put the code for policy approximation and ensemble online at https://www.dropbox.com/s/jlc6dtxo580lpl2/maddpg_ensemble_and_approx_code.zip?dl=0