Abstract
Combinatorial optimization problems such as the determination of the ground-state energy of Ising spin-glasses, the construction of low-autocorrelation binary sequences, and the traveling salesman problem are analyzed with use of zero-temperature scaling concepts. The difficulty of obtaining good ground states with most heuristic algorithms is influenced by the number of local minima. It is shown that the number of such minima is related to the thermal eigenvalue at the zero-temperature fixed point.

This publication has 21 references indexed in Scilit: