DocumentCode :
1611871
Title :
Kernel Isomap on Noisy Manifold
Author :
Choi, Heeyoul ; Choi, Seungjin
Author_Institution :
Dept. of Comput. Sci., Pohang Univ. of Sci. & Technol.
fYear :
2005
Firstpage :
208
Lastpage :
213
Abstract :
In the human brain, it is well known that perception is based on similarity rather than coordinates and it is carried out on the manifold of data set. Isomap (Tenenbaum et al., 2000) is one of widely-used low-dimensional embedding methods where approximate geodesic distance on a weighted graph is used in the framework of classical scaling (metric MDS). In this paper, we consider two critical issues missing in Isomap: (1) generalization property; (2) topological stability and present our robust kernel Isomap method, armed with such two properties. The useful behavior and validity of our robust kernel Isomap, is confirmed through numerical experiments with several data sets including real world data
Keywords :
brain; generalisation (artificial intelligence); graph theory; visual perception; classical scaling; generalization property; geodesic distance; human brain; kernel Isomap; metric multidimensional scaling; noisy manifold; topological stability; weighted graph; Humans; Information retrieval; Kernel; Machine learning; Manifolds; Principal component analysis; Robots; Robust stability; Robustness; Training data;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Development and Learning, 2005. Proceedings., The 4th International Conference on
Conference_Location :
Osaka
Print_ISBN :
0-7803-9226-4
Type :
conf
DOI :
10.1109/DEVLRN.2005.1490986
Filename :
1490986
Link To Document :
بازگشت