OPCC benchmarks
收藏arXiv2022-05-22 更新2024-06-21 收录
下载链接：
https://github.com/koulanurag/opcc
下载链接
链接失效反馈官方服务：
资源简介：
OPCC benchmarks是由俄勒冈州立大学电气工程与计算机科学学院的研究人员创建的一系列数据集，旨在评估离线策略比较的置信度。该数据集包含7个不同的环境数据集，用于离线强化学习。数据集通过添加策略比较查询集合来构建，以评估策略在不同世界状态下的性能。这些数据集公开可用，旨在支持社区对离线策略比较置信度评估的研究和扩展。
OPCC benchmarks are a suite of datasets created by researchers from the School of Electrical Engineering and Computer Science at Oregon State University, designed to evaluate the confidence of offline policy comparison. This suite includes 7 distinct environment datasets for offline reinforcement learning. These datasets are constructed by integrating a set of policy comparison queries to assess the performance of policies across diverse world states. All datasets are publicly available, with the goal of supporting the research and advancement of community studies on confidence evaluation for offline policy comparison.
提供机构：
俄勒冈州立大学电气工程与计算机科学学院
创建时间：
2022-05-22
