English

Efficient Inverse Maintenance and Faster Algorithms for Linear Programming

Data Structures and Algorithms 2015-10-15 v3 Numerical Analysis Optimization and Control

Abstract

In this paper, we consider the following inverse maintenance problem: given ARn×dA \in \mathbb{R}^{n\times d} and a number of rounds rr, we receive a n×nn\times n diagonal matrix D(k)D^{(k)} at round kk and we wish to maintain an efficient linear system solver for ATD(k)AA^{T}D^{(k)}A under the assumption D(k)D^{(k)} does not change too rapidly. This inverse maintenance problem is the computational bottleneck in solving multiple optimization problems. We show how to solve this problem with O~(nnz(A)+dω)\tilde{O}(nnz(A)+d^{\omega}) preprocessing time and amortized O~(nnz(A)+d2)\tilde{O}(nnz(A)+d^{2}) time per round, improving upon previous running times for solving this problem. Consequently, we obtain the fastest known running times for solving multiple problems including, linear programming and computing a rounding of a polytope. In particular given a feasible point in a linear program with dd variables, nn constraints, and constraint matrix ARn×dA\in\mathbb{R}^{n\times d}, we show how to solve the linear program in time O~(nnz(A)+d2)dlog(ϵ1))\tilde{O}(nnz(A)+d^{2})\sqrt{d}\log(\epsilon^{-1})). We achieve our results through a novel combination of classic numerical techniques of low rank update, preconditioning, and fast matrix multiplication as well as recent work on subspace embeddings and spectral sparsification that we hope will be of independent interest.

Keywords

Cite

@article{arxiv.1503.01752,
  title  = {Efficient Inverse Maintenance and Faster Algorithms for Linear Programming},
  author = {Yin Tat Lee and Aaron Sidford},
  journal= {arXiv preprint arXiv:1503.01752},
  year   = {2015}
}

Comments

In an older version of this paper, we mistakenly claimed an improved running time for Dikin walk by noting solely the improved running time for linear system solving and ignoring the determinant computation

R2 v1 2026-06-22T08:45:32.320Z