Experimental implementation of local adiabatic evolution algorithms by an NMR quantum information processor

Abstract

Quantum adiabatic algorithm is a method of solving computational problems by evolving the ground state of a slowly varying Hamiltonian. The technique uses evolution of the ground state of a slowly varying Hamiltonian to reach the required output state. In some cases, such as the adiabatic versions of Grover's search algorithm and Deutsch-Jozsa algorithm, applying the global adiabatic evolution yields a complexity similar to their classical algorithms. However, using the local adiabatic evolution, the algorithms given by J. Roland and N. J. Cerf for Grover's search [ Phys. Rev. A. {\bf 65} 042308(2002)] and by Saurya Das, Randy Kobes and Gabor Kunstatter for the Deutsch-Jozsa algorithm [Phys. Rev. A. {\bf 65}, 062301 (2002)], yield a complexity of order $\sqrt{N}$ (where N=2$^{\rm n}$ and n is the number of qubits). In this paper we report the experimental implementation of these local adiabatic evolution algorithms on a two qubit quantum information processor, by Nuclear Magnetic Resonance.