Overview - 2023.2 English

Vitis Libraries

Release Date
2023.2 English

In probability theory and statistics, a covariance matrix, (aka, variance-covariance matrix) is a :math: Ntimes N square matrix that contains the variances and covariances associated with N observed variables. The diagonal elements of the matrix contain the variances of the variables, and the off-diagonal elements contain the covariances between all possible pairs of variables. At the same time, in order to solve an ill-posed problem or to prevent overfitting, there are four ways to regularize the covariance matrix, including hard-thresholding. soft-thresholding, banding, and tapering.