search

yalidu / liir

Learning Individual Intrinsic Reward in MARL
https://papers.nips.cc/paper/8691-liir-learning-individual-intrinsic-reward-in-multi-agent-reinforcement-learning
63 stars 19 forks source link

Where is the codes for Visualizing the Learned Intrinsic Reward? #6

Closed GuiyangLuo closed 4 years ago

GuiyangLuo commented 4 years ago

Hi,

Thanks for your awesome work on MARL.

Where are the codes for Visualizing the Learned Intrinsic Reward? Did you Visualize it by Replay? How to connect it with intrinsic reward?

yalidu commented 4 years ago

In the testing time, you can pause after each step to save the intrinsic reward and the visualized image.

GuiyangLuo commented 4 years ago

Thanks so much. By the way, the LIIR algorithm achieves a poor performance on the hard scenario (e.g., 5m_vs_6m). Does it have a higher variance?

yalidu commented 4 years ago

I did not test on the antisymmetric scenarios. There is another paper applying LIIR to other different maps. https://arxiv.org/pdf/2007.02529.pdf

GuiyangLuo commented 4 years ago

Appreciate it so much! Thanks

luoguiyang111@gmail.com

From: Yali Du Date: 2020-09-16 14:49 To: yalidu/liir CC: GuiyangLuo; Author Subject: Re: [yalidu/liir] Where is the codes for Visualizing the Learned Intrinsic Reward? (#6) I did not test on the antisymmetric scenarios. There is another paper applying LIIR to other different maps. https://arxiv.org/pdf/2007.02529.pdf — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub, or unsubscribe.