共 3 条
- [1] Bai Z., 2000, TEMPLATES SOLUTION A, DOI DOI 10.1137/1.9780898719581
- [2] BENGIO Y, 2004, ADV NEURAL INFORMATI
- [3] SAUL LK, 2003, J MACHINE LEARNING R, V4, P119
Incremental locally linear embedding
被引：102
作者：
Kouropteva, O
Okun, O
Pietikäinen, M
机构：
[1] Univ Oulu, Machine Vis Grp, Infotech Oulu, Oulu 90014, Finland
[2] Univ Oulu, Dept Elect & Informat Engn, Oulu 90014, Finland
关键词：
dimensionality reduction;
LLE;
online mapping;
topology preservation;
D O I：
10.1016/j.patcog.2005.04.006
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ;
0812 ;
0835 ;
1405 ;
摘要：
The locally linear embedding (LLE) algorithm belongs to a group of manifold learning methods that not only merely reduce data dimensionality, but also attempt to discover a true low dimensional structure of the data. In this paper, we propose an incremental version of LLE and experimentally. demonstrate its advantages in terms of topology preservation. Also compared to the original (batch) LLE, the incremental LLE needs to solve a much smaller optimization problem. (c) 2005 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved.
引用
收藏
页码：1764 / 1767
页数：4
相关论文