papers
papers copied to clipboard
upura
SEERL: Sample Efficient Ensemble Reinforcement Learning
どんなもの？
機械学習の教師あり学習では一般的な「アンサンブル」を、強化学習の文脈で使う話。「適切に多様なポリシー設計」が必要らしく、難しそう。
https://arxiv.org/abs/2001.05209
Hi @upura ! It seems you are also interested in this paper. Have you ever found the related code of this paper? I am also interesting in it. Thanks!
@pengzhenghao I'm not familiar to this field, but how is it? https://paperswithcode.com/paper/sample-efficient-reinforcement-learning-with
Well thanks! But I am afraid the link presented in the paperswithcode points to a wrong repo, which is the tensorflow model garden but not the code for the specified paper.
In fact, I have searched the github with the paper's title but find nothing. It seems that the author of the paper do not released the code currently. Thanks again!
By the way, I read the translation of your Japanese comment on the paper. Do you suggest that the paper does a great job on finding the diverse policies or you suggest that the paper do not provide satisfied solution in finding the diverse policies?