Title :
The self-organizing reduced kernel density estimator
Author :
Holmström, Lasse ; Hämäläinen, Ari
Author_Institution :
Rolf Nevanlinna Inst., Helsinki Univ., Finland
Abstract :
The problem of reducing the number of terms in a kernal density estimator is considered. An online algorithm is proposed, where the kernel centers are tuned using the self-organizing feature map. The performance of the algorithm is compared with standard methods in numerical simulations
Keywords :
self-organising feature maps; kernal density estimator; kernel centers; numerical simulations; self-organizing feature map; Clustering algorithms; Kernel; Nearest neighbor searches; Neural networks; Numerical simulation; Pattern recognition; Probability density function; Smoothing methods; Statistical distributions;
Conference_Titel :
Neural Networks, 1993., IEEE International Conference on
Conference_Location :
San Francisco, CA
Print_ISBN :
0-7803-0999-5
DOI :
10.1109/ICNN.1993.298593
