Learning to reason about contextual knowledge for planning under uncertainty
Learning to reason about contextual knowledge for planning under uncertainty
复制标题
DOI：
--
复制
发表时间：
2023
期刊：
影响因子：
--
通讯作者：
S. Amiri
中科院分区：
文献类型：
--
作者：
S. Amiri
顺序决策（SDM）方法使AI代理能够在不确定性下计算实现长期目标的行动策略。现有的研究表明，声明形式的上下文知识可以用于提高SDM方法的性能。然而，人们所获得的背景知识往往是不完整的，有时甚至是不准确的，这极大地限制了基于知识的SDM方法的适用性。在本文中，我们开发了一种新的算法，rithm基于知识的SDM，称为危险，从交互经验的原因上下文知识，适用于城市驾驶场景。实验已经进行了使用CARLA，广泛使用的自动驾驶模拟器。结果表明，与现有的基于知识的SDM基线相比，PERIL的优越性。
Sequential decision-making (SDM) methods enable AI agents to compute an action policy toward achieving long-term goals under uncertainty. Existing research has shown that contextual knowledge in declarative forms can be used for improving the performance of SDM methods. However, the contextual knowledge from people tends to be incomplete and sometimes inaccurate, which greatly limits the applicability of knowledge-based SDM methods. In this paper, we develop a novel algo-rithm for knowledge-based SDM, called PERIL, that learns from interaction experience to reason about contextual knowledge, as applied to urban driving scenarios. Experiments have been conducted using CARLA, a widely used autonomous driving simulator. Results demonstrate PERIL’s superiority in comparison to existing knowledge-based SDM baselines.