Fast Stochastic Algorithms for SVD and PCA: Convergence Properties and Convexity
Proceedings of The 33rd International Conference on Machine Learning, PMLR 48:248-256, 2016.
We study the convergence properties of the VR-PCA algorithm introduced by (Shamir, 2015) for fast computation of leading singular vectors. We prove several new results, including a formal analysis of a block version of the algorithm, and convergence from random initialization. We also make a few observations of independent interest, such as how pre-initializing with just a single exact power iteration can significantly improve the analysis, and what are the convexity and non-convexity properties of the underlying optimization problem.