push-r-bump
收藏arXiv2025-09-30 收录
下载链接:
https://github.com/zhihanyang2022/off-policy-continuous-control
下载链接
链接失效反馈官方服务:
资源简介:
该数据集的任务是在稀疏奖励环境中进行强化学习,其中没有一个算法能够始终达到100%的成功率,这需要系统性的探索。此外,在这项任务中,RDPG、RTD3和RSAC算法都展示了一些成功的案例。
The task of this dataset is reinforcement learning in sparse-reward environments, where no algorithm can consistently attain a 100% success rate, thus requiring systematic exploration. Furthermore, algorithms including RDPG, RTD3 and RSAC have demonstrated some successful cases in this task.
提供机构:
Authors' implementation



