A new lazy-update algorithm claims near-linear O(ε^{-1}nd log(n/d)) time for approximate John ellipsoids, but key proof steps and complexity accounting contain gaps.
On computing approximate Lewis weights
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
In this note we provide and analyze a simple method that given an $n \times d$ matrix, outputs approximate $\ell_p$-Lewis weights, a natural measure of the importance of the rows with respect to the $\ell_p$ norm, for $p \geq 2$. More precisely, we provide a simple post-processing procedure that turns natural one-sided approximate $\ell_p$-Lewis weights into two-sided approximations. When combined with a simple one-sided approximation algorithm presented by Lee (PhD thesis, `16) this yields an algorithm for computing two-sided approximations of the $\ell_p$-Lewis weights of an $n \times d$-matrix using $\mathrm{poly}(d,p)$ approximate leverage score computations. While efficient high-accuracy algorithms for approximating $\ell_p$-Lewis had been established previously by Fazel, Lee, Padmanabhan and Sidford (SODA `22), the simple structure and approximation tolerance of our algorithm may make it of use for different applications.
fields
cs.DS 1years
2025 1verdicts
REJECT 1representative citing papers
citing papers explorer
-
John Ellipsoids via Lazy Updates
A new lazy-update algorithm claims near-linear O(ε^{-1}nd log(n/d)) time for approximate John ellipsoids, but key proof steps and complexity accounting contain gaps.