TY - GEN
T1 - Fast non-linear dimension reduction
AU - Kambhatla, Nandakishore
AU - Leen, Todd K.
PY - 1993/1/1
Y1 - 1993/1/1
N2 - This paper presents a new algorithm for nonlinear dimension reduction. The algorithm builds a piece-wise linear model of the data. This piece-wise linear model provides compression that is superior to the globally linear model produced by principal component analysis. On several examples the piece-wise linear model also provides compression that is superior to the global non-linear model constructed by a five-layer, autoassociative neural network. Furthermore, the new algorithm trains significantly faster than the autoassociative network.
AB - This paper presents a new algorithm for nonlinear dimension reduction. The algorithm builds a piece-wise linear model of the data. This piece-wise linear model provides compression that is superior to the globally linear model produced by principal component analysis. On several examples the piece-wise linear model also provides compression that is superior to the global non-linear model constructed by a five-layer, autoassociative neural network. Furthermore, the new algorithm trains significantly faster than the autoassociative network.
UR - http://www.scopus.com/inward/record.url?scp=84943245912&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84943245912&partnerID=8YFLogxK
U2 - 10.1109/ICNN.1993.298730
DO - 10.1109/ICNN.1993.298730
M3 - Conference contribution
AN - SCOPUS:84943245912
T3 - IEEE International Conference on Neural Networks - Conference Proceedings
SP - 1213
EP - 1218
BT - 1993 IEEE International Conference on Neural Networks, ICNN 1993
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - IEEE International Conference on Neural Networks, ICNN 1993
Y2 - 28 March 1993 through 1 April 1993
ER -