English

Accelerated Approximate Optimization of Multi-Commodity Flows on Directed Graphs

Data Structures and Algorithms 2025-04-01 v1 Optimization and Control

Abstract

We provide m1+o(1)kϵ1m^{1+o(1)}k\epsilon^{-1}-time algorithms for computing multiplicative (1ϵ)(1 - \epsilon)-approximate solutions to multi-commodity flow problems with kk-commodities on mm-edge directed graphs, including concurrent multi-commodity flow and maximum multi-commodity flow. To obtain our results, we provide new optimization tools of potential independent interest. First, we provide an improved optimization method for solving q,p\ell_{q, p}-regression problems to high accuracy. This method makes O~q,p(k)\tilde{O}_{q, p}(k) queries to a high accuracy convex minimization oracle for an individual block, where O~q,p()\tilde{O}_{q, p}(\cdot) hides factors depending only on qq, pp, or poly(logm)\mathrm{poly}(\log m), improving upon the O~q,p(k2)\tilde{O}_{q, p}(k^2) bound of [Chen-Ye, ICALP 2024]. As a result, we obtain the first almost-linear time algorithm that solves q,p\ell_{q, p} flows on directed graphs to high accuracy. Second, we present optimization tools to reduce approximately solving composite 1,\ell_{1, \infty}-regression problems to solving mo(1)ϵ1m^{o(1)}\epsilon^{-1} instances of composite q,p\ell_{q, p}-regression problem. The method builds upon recent advances in solving box-simplex games [Jambulapati-Tian, NeurIPS 2023] and the area convex regularizer introduced in [Sherman, STOC 2017] to obtain faster rates for constrained versions of the problem. Carefully combining these techniques yields our directed multi-commodity flow algorithm.

Keywords

Cite

@article{arxiv.2503.24373,
  title  = {Accelerated Approximate Optimization of Multi-Commodity Flows on Directed Graphs},
  author = {Li Chen and Andrei Graur and Aaron Sidford},
  journal= {arXiv preprint arXiv:2503.24373},
  year   = {2025}
}
R2 v1 2026-06-28T22:41:00.987Z