...
首页> 外文期刊>Journal of Experimental and Theoretical Artificial Intelligence >A mosaic of Chu spaces and Channel Theory Ⅱ: applications to object identification and mereological complexity
【24h】

A mosaic of Chu spaces and Channel Theory Ⅱ: applications to object identification and mereological complexity

机译:楚空间和通道理论Ⅱ的拼接:在对象识别和简写复杂性中的应用

获取原文
获取原文并翻译 | 示例
           

摘要

In this Part II of a two-part work, we proceed from the survey of concepts and techniques of Chu spaces and Channel Theory in Part I to the characterisation of human visual object identification, beginning with the construction of uncategorised object files and proceeding through categorisation, individual object identification and the tracking of object identity through time. We investigate the relationship between abstraction and mereological categorisation, particularly as these affect object-identity tracking. This we accomplish in terms of information flow that is semantically structured in terms of local logics, providing an inferential mechanism for object identification and tracking. We introduce categorical Cone-Cocone Diagrams to explicitly capture a bidirectional duality between 'token' roles and 'type' roles, and show that all representations can be considered to play both roles. We discuss the emergence of mereotopology from the representation of classifications with underlying simplicial complexes, and briefly explore the emergence of geometric relations and interactions between objects. Throughout we discuss the empirical support for the utility of this descriptive mechanism, particularly as a scale-free organisation that is applicable to cognition and to AI systems in general, and raise open issues and problems.
机译:在这个由两部分组成的工作的第二部分中,我们从对第一部分的楚空间和通道理论的概念和技术的调查开始,到对人类视觉对象识别的表征,从构建未分类的对象文件开始,然后进行分类,单个对象标识以及通过时间跟踪对象标识。我们研究抽象和单纯性分类之间的关系,尤其是当它们影响对象身份跟踪时。我们根据信息流(根据本地逻辑在语义上进行结构化)来完成此操作,从而为对象识别和跟踪提供了一种推理机制。我们引入了分类锥锥图,以明确地捕获“令牌”角色和“类型”角色之间的双向对偶,并表明所有表示形式都可以考虑同时扮演这两个角色。我们从具有基本简单复杂性的分类表示中讨论了光字拓扑的出现,并简要探讨了几何关系和对象之间相互作用的出现。在整个过程中,我们讨论了对这种描述机制的效用的经验支持,尤其是作为无规模组织,适用于认知和整个AI系统,并提出了未解决的问题。

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