HIGHLIGHTS
SUMMARY
The aim is to test for association between two sets of variables X and Y, such as microbiome composition, gene_expression, or profiles of other types of genomic data. The Hilbert-Schmidt independence criterion (HSIC) was first proposed by Gretton et_al20. When characteristic kernels, such as the Gaussian kernel or Laplacian kernel, are used for kX and kY, HSIC(fXY )=0 if and only if fXY=fX fY. An empirical estimate of HSIC was proposed by Gretton et_al20: HSIC= N N N k X, X k Y, Y + k X, X k Y, Y - kX . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.