Search papers, labs, and topics across Lattice.
The msPCA R package introduces an efficient method for performing sparse principal component analysis (PCA) with multiple components, utilizing an alternating maximization algorithm. This approach allows for the generation of sparse loading vectors that explain significant variance in high-dimensional datasets while ensuring non-redundancy through orthogonality or zero pairwise correlation. Benchmarks demonstrate that msPCA effectively handles sparse PCA problems with thousands of features, achieving competitive runtimes and maintaining a high fraction of explained variance.
Achieving sparse PCA with thousands of features while ensuring non-redundancy is now feasible and efficient with the msPCA package.
We present msPCA: an open-source R package for sparse principal component analysis with multiple components. It implements an alternating maximization algorithm to generate a set of sparse loading vectors that collectively explain a large fraction of the variance in a dataset, while remaining non-redundant. The algorithm supports two definitions of non-redundancy: either orthogonality of the loading vectors or zero pairwise correlation between principal components (PCs). In the reported benchmarks, msPCA solves sparse PCA problems with thousands of features, achieving competitive runtimes while producing sparse components with controlled feasibility violations and a high fraction of variance explained.
