English

Composable Core-sets for Determinant Maximization Problems via Spectral Spanners

Data Structures and Algorithms 2019-11-19 v2 Machine Learning Machine Learning

Abstract

We study a spectral generalization of classical combinatorial graph spanners to the spectral setting. Given a set of vectors VdV\subseteq \Re^d, we say a set UVU\subseteq V is an α\alpha-spectral spanner if for all vVv\in V there is a probability distribution μv\mu_v supported on UU such that vvαEuμvuu.vv^\intercal \preceq \alpha\cdot\mathbb{E}_{u\sim\mu_v} uu^\intercal. We show that any set VV has an O~(d)\tilde{O}(d)-spectral spanner of size O~(d)\tilde{O}(d) and this bound is almost optimal in the worst case. We use spectral spanners to study composable core-sets for spectral problems. We show that for many objective functions one can use a spectral spanner, independent of the underlying functions, as a core-set and obtain almost optimal composable core-sets. For example, for the determinant maximization problem we obtain an O~(k)k\tilde{O}(k)^k-composable core-set and we show that this is almost optimal in the worst case. Our algorithm is a spectral analogue of the classical greedy algorithm for finding (combinatorial) spanners in graphs. We expect that our spanners find many other applications in distributed or parallel models of computation. Our proof is spectral. As a side result of our techniques, we show that the rank of diagonally dominant lower-triangular matrices are robust under `small perturbations' which could be of independent interests.

Keywords

Cite

@article{arxiv.1807.11648,
  title  = {Composable Core-sets for Determinant Maximization Problems via Spectral Spanners},
  author = {Piotr Indyk and Sepideh Mahabadi and Shayan Oveis Gharan and Alireza Rezaei},
  journal= {arXiv preprint arXiv:1807.11648},
  year   = {2019}
}

Comments

To appear in SODA 2020

R2 v1 2026-06-23T03:19:55.224Z