arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2410.09466cs.LG
收藏

双曲空间中的强化学习:模型与实验

Reinforcement Learning in Hyperbolic Spaces: Models and Experiments

Faculty of Natural Sciences and Mathematics(自然科学与数学系);University of Montenegro(黑山大学);Faculty of Engineering(工程学院);University of Rijeka(里耶卡大学);Faculty of Technical Engineering(技术工程学院);University of Bihać(比哈奇大学)

Vladimir Jaćimović, Zinaid Kapić, Aladin Crnkić

AI总结:

本文将多智能体未知环境探索形式化为双曲空间中的强化学习问题,通过引入统计与动力学模型并基于黑盒优化视角实现相关算法。

摘要:

我们考察了智能体(或两个智能体)在无任何先验信息下探索未知环境的五种设置。尽管看似截然不同,但它们均可形式化为双曲空间中的强化学习(RL)问题。更准确地说,将双曲度量赋予动作空间是很自然的。我们引入了解决此类问题所需的统计和动力学模型,并实现了基于该框架的算法。在整篇论文中,我们通过黑盒优化的视角来审视RL。

英文摘要:

We examine five setups where an agent (or two agents) seeks to explore unknown environment without any prior information. Although seemingly very different, all of them can be formalized as Reinforcement Learning (RL) problems in hyperbolic spaces. More precisely, it is natural to endow the action spaces with the hyperbolic metric. We introduce statistical and dynamical models necessary for addressing problems of this kind and implement algorithms based on this framework. Throughout the paper we view RL through the lens of the black-box optimization.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑