Goto

Collaborating Authors

 capped msg

Stochastic Optimization of PCA with Capped MSG

Neural Information Processing Systems

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. Papers published at the Neural Information Processing Systems Conference.

Stochastic Optimization of PCA with Capped MSG

arXiv.org Machine Learning

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.