Some approaches to analogical mapping with structure-sensitive distributed representations

作     者：Rachkovskij, DA 

作者机构：VM Glushkov Cybernet Ctr UA-03680 Kiev Ukraine 

出 版 物：《JOURNAL OF EXPERIMENTAL & THEORETICAL ARTIFICIAL INTELLIGENCE》 (人工智能实验与理论杂志)

年 卷 期：2004年第16卷第3期

页      面：125-144页

核心收录：

学科分类：12[管理学] 1201[管理学-管理科学与工程(可授管理学、工学学位)] 081104[工学-模式识别与智能系统] 08[工学] 0835[工学-软件工程] 0811[工学-控制科学与工程] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

主　　题：analogy analogical mapping reasoning connectionist symbol processing distributed representations representation of structure codevector sparse coding binary coding binding associative-projective neural networks APNN 

摘      要：This paper presents some techniques for analogical mapping using associative-projective neural networks (APNNs). Sparse binary distributed representations of constant high dimensionality are constructed on-the-fly for hierarchical structures of various complexity. Such representations encode both surface and structural similarity of analogical episodes. The introduced mapping approaches are illustrated using test analogies.