English

Incremental Shortest Paths in Almost Linear Time via a Modified Interior Point Method

Data Structures and Algorithms 2025-06-25 v1

Abstract

We give an algorithm that takes a directed graph GG undergoing mm edge insertions with lengths in [1,W][1, W], and maintains (1+ϵ)(1+\epsilon)-approximate shortest path distances from a fixed source ss to all other vertices. The algorithm is deterministic and runs in total time m1+o(1)logWm^{1+o(1)}\log W, for any ϵ>exp((logm)0.99)\epsilon > \exp(-(\log m)^{0.99}). This is achieved by designing a nonstandard interior point method to crudely detect when the distances from ss other vertices vv have decreased by a (1+ϵ)(1+\epsilon) factor, and implementing it using the deterministic min-ratio cycle data structure of [Chen-Kyng-Liu-Meierhans-Probst, STOC 2024].

Keywords

Cite

@article{arxiv.2506.19207,
  title  = {Incremental Shortest Paths in Almost Linear Time via a Modified Interior Point Method},
  author = {Yang P. Liu},
  journal= {arXiv preprint arXiv:2506.19207},
  year   = {2025}
}

Comments

20 pages

R2 v1 2026-07-01T03:30:35.906Z