English

Karp's patching algorithm on dense digraph

Combinatorics 2025-05-29 v1 Discrete Mathematics

Abstract

We consider the following question. We are given a dense digraph DD with nn vertices and minimum in- and out-degree at least αn\alpha n, where α>1/2\alpha>1/2 is a constant. The edges E(D)E(D) of DD are given independent edge costs C(e),eE(D)C(e),e\in E(D), such that (i) CC has a density ff that satisfies f(x)=a+bx+O(x2)f(x)=a+bx+O(x^2), for constants a>0,ba>0,b as x0x\to 0 and such that in general either (ii) Pr(Cx)\ae\bx\Pr(C\geq x)\leq \a e^{-\b x} for constants \a,\b>0\a,\b>0, or f(x)=0f(x)=0 for x>\nx>\n for some constant \n>0\n>0. Let C(i,j),i,j[n]C(i,j),i,j\in[n] be the associated n×nn\times n cost matrix where C(i,j)=C(i,j)=\infty if (i,j)E(i,j)\notin E. We show that w.h.p. (a small modification to) the patching algorithm of Karp finds a tour for the asymmetric traveling salesperson problem that is asymptotically equal to that of the associated assignment problem. The algorithm runs in polynomial time.

Keywords

Cite

@article{arxiv.2505.21645,
  title  = {Karp's patching algorithm on dense digraph},
  author = {Alan Frieze},
  journal= {arXiv preprint arXiv:2505.21645},
  year   = {2025}
}

Comments

arXiv admin note: text overlap with arXiv:2209.06279