Provides methods to perform block diagonal covariance matrix detection using singular vectors ('BD-SVD'), which can be extended to inherently sparse principal component analysis ('IS-PCA'). The methods are described in Bauer (2025)
Block detection using singular value decomposition