arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2108.03213cs.LGcs.AIstat.ML
收藏

时间抽象部分模型

Temporally Abstract Partial Models

McGill University(麦吉尔大学);Mila;DeepMind

Khimya Khetarpal, Zafarali Ahmed, Gheorghe Comanici, Doina Precup

摘要:

人类和动物都有能力在许多时间尺度上对不同的行动过程进行推理和预测。在强化学习中,期权模型(Sutton,Preup\& Singh,1999; Preup,2000)为这种时间抽象的预测和推理提供了框架。自然智能体也能够将注意力集中在特定情况下相关或可行的行动方案上,有时称为负担得起的行动。在本文中,我们定义了一个概念的选项的启示,并开发时间抽象的部分选项模型,考虑到一个选项可能只在某些情况下负担得起的事实。我们分析了使用这种模型时,在规划和学习中的估计和近似误差之间的权衡,并确定了一些有趣的特殊情况。此外,我们实证证明了部分期权模型对规划效率的潜在影响。

英文摘要:

Humans and animals have the ability to reason and make predictions about different courses of action at many time scales. In reinforcement learning, option models (Sutton, Precup \& Singh, 1999; Precup, 2000) provide the framework for this kind of temporally abstract prediction and reasoning. Natural intelligent agents are also able to focus their attention on courses of action that are relevant or feasible in a given situation, sometimes termed affordable actions. In this paper, we define a notion of affordances for options, and develop temporally abstract partial option models, that take into account the fact that an option might be affordable only in certain situations. We analyze the trade-offs between estimation and approximation error in planning and learning when using such models, and identify some interesting special cases. Additionally, we demonstrate empirically the potential impact of partial option models on the efficiency of planning.

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。

↑