Approximate Carath\'eodory bounds via Discrepancy Theory

From MaRDI portal
Publication:6404344

arXiv2207.03614MaRDI QIDQ6404344FDOQ6404344


Authors: Victor Reis, Thomas Rothvoß Edit this on Wikidata


Publication date: 7 July 2022

Abstract: The approximate Carath'eodory problem in general form is as follows: Given two symmetric convex bodies P,QsubseteqmathbbRm, a parameter kinmathbbN and mathbfzinextrmconv(X) with XsubseteqP, find mathbfv1,ldots,mathbfvkinX so that |mathbfzfrac1ksumi=1kmathbfvi|Q is minimized. Maurey showed that if both P and Q coincide with the |cdot|p-ball, then an error of O(sqrtp/k) is possible. We prove a reduction to the vector balancing constant from discrepancy theory which for most cases can provide tight bounds for general P and Q. For the case where P and Q are both |cdot|p-balls we prove an upper bound of sqrtfracminp,log(frac2mk)k. Interestingly, this bound cannot be obtained taking independent random samples; instead we use the Lovett-Meka random walk. We also prove an extension to the more general case where P and Q are |cdot|p and |cdot|q-balls with 2leqpleqqleqinfty.













This page was built for publication: Approximate Carath\'eodory bounds via Discrepancy Theory

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6404344)