Inapproximability research for metric TSP

113 Views Asked by At

I'm doing research into improving the inapproximability ratio for the metric/graphic Traveling Salesman Problem. As I've been reading through the literature in this field, I've noticed that most of the papers rely on relating TSP to related combinatorial optimization problems about finding solutions to as many of a large system of linear equations as possible (like MAX-E3-LIN2). However, none of these articles seem to explain why it is better to work with TSP in this basis, or why the manipulation is useful to demonstrating inapproximability. Can anyone point me to a good book or tutorial review that explains how this technique works (perhaps the first article to employ it would be a good resource, but I haven't been able to locate it!)

1

There are 1 best solutions below

1
On BEST ANSWER

The connection between the Traveling Salesman Problem (TSP) and linear equations arises from the use of linear programming relaxations and semidefinite programming relaxations for approximating the TSP. These relaxations provide a way to obtain lower bounds on the optimal TSP tour length.

The connection between TSP and linear equations is typically established through the use of the Held-Karp relaxation, also known as the subtour elimination LP relaxation. This relaxation is obtained by relaxing the integer constraints of the TSP formulation and formulating it as a linear programming problem. The solution to this relaxation provides a lower bound on the optimal TSP tour length.

while these dont directly explain the technique advantages or superiority they could be a decent reference with far greater insights than I can put together.

Approximation Algorithms by Vijay V. Vazirani

The Design of Approximation Algorithms by David P. Williamson and David B. Shmoys

Approximation Algorithms for NP-Hard Problems by Dorit S. Hochbaum