An O(log n/log log n)-Approximation Algorithm for the Asymmetric Traveling Salesman Problem

被引:40
|
作者
Asadpour, Arash [1 ]
Goemans, Michel X. [2 ]
Madry, Aleksander [3 ]
Gharan, Shayan Oveis [4 ]
Saberi, Amin [5 ]
机构
[1] NYU, Stern Sch Business, Dept Informat Operat & Management Sci, 550 1St Ave, New York, NY 10012 USA
[2] MIT, Dept Math, Cambridge, MA 02139 USA
[3] MIT, Dept Elect Engn & Comp Sci, Cambridge, MA 02139 USA
[4] Univ Washington, Dept Comp Sci & Engn, Seattle, WA 98105 USA
[5] Stanford Univ, Dept Management Sci & Engn, Stanford, CA 94305 USA
基金
美国国家科学基金会;
关键词
traveling salesman problem; linear programming; maximum entropy; thin tree; Held-Karp relaxation; randomized rounding; APPROXIMATION ALGORITHM;
D O I
10.1287/opre.2017.1603
中图分类号
C93 [管理学];
学科分类号
12 ; 1201 ; 1202 ; 120202 ;
摘要
We present a randomized O(log n /log log n)-approximation algorithm for the asymmetric traveling salesman problem (ATSP). This provides the first asymptotic improvement over the long-standing Theta(log n)-approximation bound stemming from the work of Frieze et al. (1982) [ Frieze AM, Galbiati G, Maffioki F (1982) On the worst-case performance of some algorithms for the asymmetric traveling salesman problem. Networks 12(1): 23-39]. The key ingredient of our approach is a new connection between the approximability of the ATSP and the notion of so-called thin trees. To exploit this connection, we employ maximum entropy rounding-a novel method of randomized rounding of LP relaxations of optimization problems. We believe that this method might be of independent interest.
引用
收藏
页码:1043 / 1061
页数:19
相关论文
共 50 条