no code implementations • 6 Dec 2023 • Joonsuk Kang, Matthew Stephens
We show that this formulation also leads to a penalized decomposition of the covariance (or Gram) matrix, $\mathbf{X}^T\mathbf{X}$.