MineRL Competition Dataset
收藏arXiv2025-09-30 收录
下载链接：
https://www.aicrowd.com/challenges/neurips-2020-minerl-competition/leaderboards
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是为了评估在MineRL竞赛中强化学习和模仿学习算法的性能而设计的，特别针对获取钻石（ObtainDiamond）任务。该数据集包含了两次评估轮次的得分，以及多个剧集中的详细性能指标。规模上，有90多个团队参与了竞赛，并提交了大约500份作品。这项任务专注于对获取钻石任务的强化学习和模仿学习研究。
This dataset is designed to evaluate the performance of reinforcement learning and imitation learning algorithms in the MineRL Competition, specifically targeting the ObtainDiamond task. It includes scores from two evaluation rounds and detailed performance metrics across multiple episodes. In terms of scale, over 90 teams participated in the competition and submitted approximately 500 entries. This task focuses on research on reinforcement learning and imitation learning for the ObtainDiamond task.
提供机构：
MineRL competition organizers
