Simulated Environments
收藏arXiv2025-09-30 收录
下载链接：
http://github.com/Stanford-ILIAD/easy-active-learning
下载链接
链接失效反馈官方服务：
资源简介：
该数据集包含了多种模拟环境，旨在评估在主动奖励学习中提出的信息增益方法，以提出简单问题。这些环境具有六维状态空间和三维动作空间，模拟了机器人任务。数据集的规模覆盖了不同复杂度的多个环境，其任务专注于主动奖励学习。
This dataset comprises multiple simulated environments intended to evaluate the information gain method proposed for active reward learning when posing simple questions. Each environment features a six-dimensional state space and a three-dimensional action space, simulating robotic tasks. The dataset includes environments spanning different complexity levels, with all tasks within these environments focused on active reward learning.
提供机构：
Stanford University
