The goal of subspace learning is to find a k-dimensional subspace of Rd, such that the expected squared distance between instance vectors and the subspace is as small as possible. In this paper we study subspace learning in a partial information setting, in which the learner can only observe r≤d attributes from each instance vector. We propose several efficient algorithms for this task, and analyze their sample complexity
@article{arxiv.1402.4844,
title = {Subspace Learning with Partial Information},
author = {Alon Gonen and Dan Rosenbaum and Yonina Eldar and Shai Shalev-Shwartz},
journal= {arXiv preprint arXiv:1402.4844},
year = {2016}
}