Revision as of 07:44, 29 April 2016 edit Askerlee (talk \| contribs) 14 edits No edit summary ← Previous edit		Revision as of 07:45, 29 April 2016 edit undo Askerlee (talk \| contribs) 14 edits No edit summary Next edit →
Line 20: : <math>q_{ij} = \frac{(1 + \lVert \mathbf{y}_i - \mathbf{y}_j\rVert^2)^{-1}}{\sum_{k \neq i} (1 + \lVert \mathbf{y}_k - \mathbf{y}_i\rVert^2)^{-1}}</math> Herein a heavy-tailed [[Student-t distribution]] (with one-degree of freedom, which is ~~equivalent~~the tosame ~~the~~as a [[Cauchy distribution]]) is used to measure similarities between low-dimensional points in order to allow dissimilar objects to be modeled far apart in the map. The locations of the points <math>\mathbf{y}_i</math> in the map are determined by minimizing the (non-symmetric) [[Kullback–Leibler divergence]] of the distribution <math>Q</math> from the distribution <math>P</math>, that is:

T-distributed stochastic neighbor embedding: Difference between revisions