Distributed low rank approximation of implicit functions of a matrix

Abstract

We study distributed low rank approximation in which the matrix to be approximated is only implicitly represented across the different servers. For example, each of s servers may have an n×d matrix At, and we may be interested in computing a low rank approximation to A = (Σt=1sAt), where is a function which is applied entrywise to the matrix Σt=1sAt. We show for a wide class of functions it is possible to efficiently compute a d × d rank-k projection matrix P for which A - AP-F2 ≤ A - [A]k F2 + ϵ AF2, where AP denotes the projection of A onto the row span of P, and [A]k denotes the best rank-k approximation to A given by the singular value decomposition. The communication cost of our protocols is d·(sk=ϵ)O(1), and they succeed with high probability. Our framework allows us to efficiently compute a low rank approximation to an entry-wise softmax, to a Gaussian kernel expansion, and to M-Estimators applied entrywise (i.e., forms of robust low rank approximation). We also show that our additive error approximation is best possible, in the sense that any protocol achieving relative error for these problems requires significantly more communication. Finally, we experimentally validate our algorithms on real datasets.