Heteroscedastic Gaussian Kernel-Based Topographic Maps
نویسنده
چکیده
Several learning algorithms for topographic map formation have been introduced that adopt overlapping activa-tion regions, rather than Voronoiregions, usually in the form of kernel functions. We review and introduce a numberof fixed point rules for training homogeneous, heteroscedastic but otherwise radially-symmetric Gaussian kernel-based topographic maps, or kernel topographic maps. We compare their performance for clustering a number of realworld data sets.
منابع مشابه
Fixed point rules for heteroscedastic Gaussian kernel-based topographic map formation
Abstract— We develop a number of fixed point rules for training homogeneous, heteroscedastic but otherwise radially-symmetric Gaussian kernel-based topographic maps. We extend the batch map algorithm to the heteroscedastic case and introduce two candidates of fixed point rules for which the end-states, i.e., after the neighborhood range has vanished, are identical to the maximum likelihood Gaus...
متن کاملMaximum Likelihood Topographic Map Formation
We introduce a new unsupervised learning algorithm for kernel-based topographic map formation of heteroscedastic gaussian mixtures that allows for a unified account of distortion error (vector quantization), log-likelihood, and Kullback-Leibler divergence.
متن کاملMode estimation with topographic maps
The paper reviews thoroughly a variety of issues related to mode estimation. The potential of self-organizing maps as an approach to mode detection is inquired here. The batch version of the standard SOM and a convex adjustment of it are compared with two kernel-based learning rules, namely, the generative topographic mapping and the kernelbased maximum entropy learning rule. A strategy for mod...
متن کاملOptimal Smoothing of Kernel-Based Topographic Maps with Application to Density-Based Clustering of Shapes
A crucial issue when applying topographic maps for clustering purposes is how to select the map’s overall degree of smoothness. In this paper, we develop a new strategy for optimally smoothing, by a common scale factor, the density estimates generated by Gaussian kernel-based topographic maps. We also introduce a new representation structure for images of shapes, and a new metric for clustering...
متن کاملA Comparison of Spatial Analysis Methods for the Construction of Topographic Maps of Retinal Cell Density
Topographic maps that illustrate variations in the density of different neuronal sub-types across the retina are valuable tools for understanding the adaptive significance of retinal specialisations in different species of vertebrates. To date, such maps have been created from raw count data that have been subjected to only limited analysis (linear interpolation) and, in many cases, have been p...
متن کامل