Title :
Neural tree density estimation for novelty detection
Author :
Martinez, Dominique
Author_Institution :
Lab. d´´Anal. et d´´Archit. des Syst., CNRS, Toulouse, France
fDate :
3/1/1998 12:00:00 AM
Abstract :
In this paper, a neural competitive learning tree is introduced as a computationally attractive scheme for adaptive density estimation and novelty detection. The learning rule yields equiprobable quantization of the input space and provides an adaptive focusing mechanism capable of tracking time-varying distributions. It is shown by simulation that the neural tree performs reasonably well while being much faster than any of the other competitive learning algorithms
Keywords :
adaptive estimation; data structures; maximum entropy methods; neural nets; quantisation (signal); trees (mathematics); unsupervised learning; adaptive density estimation; adaptive multivariable histogram; competitive learning tree; equiprobable quantization; learning rule; maximum entropy; neural tree density estimation; novelty detection; sequential detection; time-varying distributions; tree data structure; unsupervised learning; Clustering algorithms; Computational modeling; Convergence; Covariance matrix; Histograms; Iterative algorithms; Lattices; Neurons; Partitioning algorithms; Quantization;
Journal_Title :
Neural Networks, IEEE Transactions on
