Learning Rational Subgoals from Demonstrations and Instructions-猫眼课题宝

DOI：

10.48550/arxiv.2303.05487

发表时间：

2023-03

期刊：

ArXiv

影响因子：

通讯作者：

中科院分区：

其他

文献类型：

作者：

Zhezheng Luo;Jiayuan Mao;Jiajun Wu;Tomas Lozano-Perez;J. Tenenbaum;L. Kaelbling

我们提出了一个学习有用的子目标的框架，支持有效的长期规划以实现新的目标。我们框架的核心是理性子目标（RSG）的集合，它们本质上是环境状态的二元分类器。 RSG 可以从弱注释数据中学习，以不分段的演示轨迹的形式，与抽象任务描述配对，这些任务描述由代理最初未知的术语组成（例如，收集木材，然后工艺船，然后跨河）。我们的框架还发现 RSG 之间的依赖关系，例如，收集木材任务对于工艺船任务来说是一个有用的子目标。给定目标描述，学习的子目标和派生的依赖关系通过将有用的子目标设置为规划器的路径点，促进现成的规划算法，例如 A* 和 RRT，这显着提高了性能时间效率。项目页面：https://rsg.csail.mit.edu

We present a framework for learning useful subgoals that support efficient long-term planning to achieve novel goals. At the core of our framework is a collection of rational subgoals (RSGs), which are essentially binary classifiers over the environmental states. RSGs can be learned from weakly-annotated data, in the form of unsegmented demonstration trajectories, paired with abstract task descriptions, which are composed of terms initially unknown to the agent (e.g., collect-wood then craft-boat then go-across-river). Our framework also discovers dependencies between RSGs, e.g., the task collect-wood is a helpful subgoal for the task craft-boat. Given a goal description, the learned subgoals and the derived dependencies facilitate off-the-shelf planning algorithms, such as A* and RRT, by setting helpful subgoals as waypoints to the planner, which significantly improves performance-time efficiency. Project page: https://rsg.csail.mit.edu

Learning Rational Subgoals from Demonstrations and Instructions

文献摘要

相似文献