Whitening transformation

A whitening transformation is a linear transformation that transforms a vector of random variables with a known covariance matrix into a set of new variables whose covariance is the identity matrix meaning that they are uncorrelated and all have variance 1. The transformation is called "whitening" because it changes the input vector into a white noise vector.

Several other transformations are closely related to whitening: 1) the decorrelation transform removes only the correlations but leaves variances intact, 2) the standardization transform sets variances to 1 but leaves correlations intact, and 3) a coloring transformation transforms a vector of white random variables into a random vector with a specified covariance matrix.

Definition

Suppose $X$ is a random (column) vector with non-singular covariance matrix $M$ and mean $0$. Then the transformation $Y=WX$ with a whitening matrix $W$ satisfying the condition $W^{T}W=M^{-1}$ yields the whitened random vector $Y$ with unit diagonal covariance.

There are infinitely many possible whitening matrices $W$ that all satisfy the above condition. Commonly used choices are $W=M^{-1/2}$ (Mahalanobis or ZCA whitening), the Cholesky decomposition of $M^{-1}$ (Cholesky whitening), or the eigen-system of $M$ (PCA whitening).

Kessy et al. (2015) demonstrate that optimal whitening transforms can be singled out by investigating the cross-covariance and cross-correlation of $X$ and $Y$. For example, the unique optimal whitening transformation achieving maximal component-wise correlation between original $X$ and whitened $Y$ is produced by the whitening matrix $W=P^{-1/2}V^{-1/2}$ where $P$ is the correlation matrix and $V$ the variance matrix.

Whitening a data matrix

Whitening a data matrix follows the same transformation as for random variables. An empirical whitening transform is obtained by estimating the covariance (e.g. by maximum likelihood) and subsequently constructing a corresponding estimated whitening matrix (e.g. by Cholesky decomposition).

See also

• Decorrelation
• Principal component analysis
• Weighted least squares

External links

• http://courses.media.mit.edu/2010fall/mas622j/whiten.pdf
• The ZCA whitening transformation. Appendix A of Learning Multiple Layers of Features from Tiny Images by A. Krizhevsky.