arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1711.11289cs.AI
收藏

学习作曲技巧

Learning to Compose Skills

Georgia Institute of Technology(佐治亚理工学院)

Himanshu Sahni, Saurabh Kumar, Farhan Tejani, Charles Isbell

摘要:

我们提出了一个可区分的框架,能够学习各种各样的简单的政策,我们称之为技能的组成。通过递归地组合技能,我们可以创建显示复杂行为的层次结构。训练技能网络以生成技能状态嵌入,这些嵌入作为可训练组合函数的输入提供,组合函数进而输出整个任务的策略。我们在由多个收集和逃避任务组成的环境中进行的实验表明,这种架构能够从简单的任务中快速构建复杂的技能。此外,学习的组合功能显示了一些转移到看不见的技能组合,允许zero-shot概括。

英文摘要:

We present a differentiable framework capable of learning a wide variety of compositions of simple policies that we call skills. By recursively composing skills with themselves, we can create hierarchies that display complex behavior. Skill networks are trained to generate skill-state embeddings that are provided as inputs to a trainable composition function, which in turn outputs a policy for the overall task. Our experiments on an environment consisting of multiple collect and evade tasks show that this architecture is able to quickly build complex skills from simpler ones. Furthermore, the learned composition function displays some transfer to unseen combinations of skills, allowing for zero-shot generalizations.

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。

↑