Large-Scale Paralleled Sparse Principal Component Analysis
This work provides a faster method for researchers and practitioners analyzing large multivariate datasets with SPCA, though it is incremental as it adapts an existing approach to GPU hardware.
The paper tackled the challenge of making sparse principal component analysis (SPCA) more efficient for large datasets by developing a parallel GPU implementation of the generalized power method, achieving speedups of up to 11 times compared to CPU and 107 times compared to MatLab.
Principal component analysis (PCA) is a statistical technique commonly used in multivariate data analysis. However, PCA can be difficult to interpret and explain since the principal components (PCs) are linear combinations of the original variables. Sparse PCA (SPCA) aims to balance statistical fidelity and interpretability by approximating sparse PCs whose projections capture the maximal variance of original data. In this paper we present an efficient and paralleled method of SPCA using graphics processing units (GPUs), which can process large blocks of data in parallel. Specifically, we construct parallel implementations of the four optimization formulations of the generalized power method of SPCA (GP-SPCA), one of the most efficient and effective SPCA approaches, on a GPU. The parallel GPU implementation of GP-SPCA (using CUBLAS) is up to eleven times faster than the corresponding CPU implementation (using CBLAS), and up to 107 times faster than a MatLab implementation. Extensive comparative experiments in several real-world datasets confirm that SPCA offers a practical advantage.