Abstract
Quantum annealing is a generic algorithm for optimization problems. In contrast to simulated annealing, which uses thermal fluctuations to explore the phase space, quantum annealing exploits quantum fluctuations to drive the system from state to state. After a brief review of the basic idea and numerical results, we present our new theorems which guarantee convergence toward the optimal state under appropriate annealing schedules. In particular, stochastic implementations of quantum annealing, suitable for classical computers, will be discussed in detail. It turns out that faster annealing schedules than the case of simulated annealing suffice for convergence. This result serves as a theoretical background for faster convergence properties observed in many numerical investigations.
Users
Please
log in to take part in the discussion (add own reviews or comments).