【24h】

Embodied AI: symbol grounding through imagination

机译:体现了AI:通过想象力接地的象征

获取原文

摘要

This paper presents the principle of symbol grounding in robots. If a robot does not understand words, the robot cannot autonomously talk in the physical world. Understanding language symbols is a necessary condition for autonomous conversations. First, how humans understand language symbols is briefly studied. Humans' symbol manipulation is based on imagination. Recent brain science shows that imagination is a virtual bodily movement. Therefore, humans' symbol manipulations are based on virtual bodily movements. Based on virtual bodily movements, a principle of symbol manipulation (grounding) in robots is presented. The idea is called Embodied AI. The semantics is called the embodied semantics, and the architecture is called the embodied architecture. In the embodied AI, abstract symbols are grounded through metaphors.
机译:本文呈现了机器人符号接地的原理。如果机器人不了解单词,那么机器人就无法在物理世界中自主地谈论。理解语言符号是自主话的必要条件。首先,简要研究人类了解语言符号。人类的象征操纵是基于想象力。最近的脑科学表明,想象力是虚拟的身体运动。因此,人类的象征操纵基于虚拟体移动。基于虚拟体移动,呈现了机器人中的符号操作原理(接地)。这个想法被称为体现ai。语义被称为体现语义,并且该体系结构称为体现体系结构。在体现的AI中,抽象符号通过隐喻接地。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号