Stochastic Optimization of PCA with Capped MSG

Part of Advances in Neural Information Processing Systems 26 (NIPS 2013)

Bibtex Metadata Paper Reviews Supplemental


Raman Arora, Andy Cotter, Nati Srebro


We study PCA as a stochastic optimization problem and propose a novel stochastic approximation algorithm which we refer to as Matrix Stochastic Gradient'' (MSG), as well as a practical variant, Capped MSG. We study the method both theoretically and empirically. "