Exact solutions for recursive principal components analysis of sequences and trees

被引：0
作者：
Sperduti, Alessandro [1 ]
机构：
[1] Univ Padua, Dept Pure & Appl Math, Padua, Italy
来源：
ARTIFICIAL NEURAL NETWORKS - ICANN 2006, PT 1 | 2006年 / 4131卷
关键词：
D O I：
暂无
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
We show how a family of exact solutions to the Recursive Principal Components Analysis learning problem can be computed for sequences and tree structured inputs. These solutions are derived from eigenanalysis of extended vectorial representations of the input structures and substructures. Experimental results performed on sequences and trees generated by a context-free grammar show the effectiveness of the proposed approach.
引用
收藏
页码：349 / 356
页数：8
相关论文
共 6 条
[1]  
Callan R. E., 1997, Connection Science, V9, P139, DOI 10.1080/095400997116667
[2]   RECURSIVE DISTRIBUTED REPRESENTATIONS [J].
POLLACK, JB .
ARTIFICIAL INTELLIGENCE, 1990, 46 (1-2) :77-105
[3]  
Sperduti A., 1994, Connection Science, V6, P429, DOI 10.1080/09540099408915733
[4]   Supervised neural networks for the classification of structures [J].
Sperduti, A ;
Starita, A .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1997, 8 (03) :714-735
[5]   Recursive principal components analysis [J].
Voegtlin, T .
NEURAL NETWORKS, 2005, 18 (08) :1051-1063
[6]   Linear recursive distributed representations [J].
Voegtlin, T ;
Dominey, PF .
NEURAL NETWORKS, 2005, 18 (07) :878-895
1