Riemannian adaptive stochastic gradient algorithms on matrix manifolds

arXiv1902.01144MaRDI QIDQ6313564

Author name not available (Why is that?)

Publication date: 4 February 2019

Abstract: Adaptive stochastic gradient algorithms in the Euclidean space have attracted much attention lately. Such explorations on Riemannian manifolds, on the other hand, are relatively new, limited, and challenging. This is because of the intrinsic non-linear structure of the underlying manifold and the absence of a canonical coordinate system. In machine learning applications, however, most manifolds of interest are represented as matrices with notions of row and column subspaces. In addition, the implicit manifold-related constraints may also lie on such subspaces. For example, the Grassmann manifold is the set of column subspaces. To this end, such a rich structure should not be lost by transforming matrices to just a stack of vectors while developing optimization algorithms on manifolds. We propose novel stochastic gradient algorithms for problems on Riemannian matrix manifolds by adapting the row and column subspaces of gradients. Our algorithms are provably convergent and they achieve the convergence rate of order $mathcalO(log(T)/sqrtT)$, where $T$ is the number of iterations. Our experiments illustrate the efficacy of the proposed algorithms on several applications.

Has companion code repository: https://github.com/hiroyuki-kasai/RSOpt

This page was built for publication: Riemannian adaptive stochastic gradient algorithms on matrix manifolds

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6313564)