search

RITCHIEHuang / MAGAIL

Pytorch implementation of Multi-Agent Generative Adversarial Imitation Learning
34 stars 4 forks source link

>Realization for obtaining expert tracks #4

Open yuanyaaa opened 2 years ago

yuanyaaa commented 2 years ago

“Obtain data from the official project”, can I understand that there is no code realized to obtain expert tracks in the project？

Another question is whether the code implements the decentralized discriminator method?

Thank you for your reply

Originally posted by @yuanyaaa in https://github.com/RITCHIEHuang/MAGAIL/issues/3#issuecomment-1182904177

RITCHIEHuang commented 2 years ago

For Q1, its yes. Actually, the expert data should be collected by your own.

For Q2, what you mean 'decentralized'? Can I understand that training a unique discriminator for each agent instead of a combined one?

获取 Outlook for iOShttps://aka.ms/o0ukef

发件人: yuanya @.> 发送时间: Friday, July 15, 2022 11:59:04 AM 收件人: RITCHIEHuang/MAGAIL @.> 抄送: Subscribed @.***> 主题: [RITCHIEHuang/MAGAIL] >Realization for obtaining expert tracks (Issue #4)

“Obtain data from the official project”, can I understand that there is no code realized to obtain expert tracks in the project？

Another question is whether the code implements the decentralized discriminator method?

Thank you for your reply

Originally posted by @yuanyaaahttps://github.com/yuanyaaa in #3 (comment)https://github.com/RITCHIEHuang/MAGAIL/issues/3#issuecomment-1182904177

― Reply to this email directly, view it on GitHubhttps://github.com/RITCHIEHuang/MAGAIL/issues/4, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AD5GQAYGEJ3ROJDYDAQIS3LVUDOYRANCNFSM53UIU63A. You are receiving this because you are subscribed to this thread.Message ID: @.***>

yuanyaaa commented 2 years ago

Yes， training a unique discriminator for each agent instead of a combined one. I do not find the decentralized setting.