Sparse Principal Component Analysis via Variable Projection
This provides a more interpretable and robust technique for low-rank data analysis, particularly for large-scale or corrupted datasets, though it appears incremental as an algorithmic improvement to existing SPCA methods.
The authors tackled the problem of sparse principal component analysis (SPCA) by formulating it as a value-function optimization problem, resulting in a robust and scalable algorithm that achieves exceptional computational efficiency and diagnostic performance on both synthetic and real-world data.
Sparse principal component analysis (SPCA) has emerged as a powerful technique for modern data analysis, providing improved interpretation of low-rank structures by identifying localized spatial structures in the data and disambiguating between distinct time scales. We demonstrate a robust and scalable SPCA algorithm by formulating it as a value-function optimization problem. This viewpoint leads to a flexible and computationally efficient algorithm. Further, we can leverage randomized methods from linear algebra to extend the approach to the large-scale (big data) setting. Our proposed innovation also allows for a robust SPCA formulation which obtains meaningful sparse principal components in spite of grossly corrupted input data. The proposed algorithms are demonstrated using both synthetic and real world data, and show exceptional computational efficiency and diagnostic performance.