Computer Science · Artificial Intelligence
TRANSFORM TO FIND EFFECTIVE HAMILTONIANS
Michael James O’hara, Doctor Of Philosophy, Michael James O’hara
See on the mapAbstract
This thesis explores two mathematical aspects of adiabatic quantum computa-tion. Adiabatic quantum computation depends on the adiabatic theorem of quantum mechanics, and (a) we provide a rigorous formulation of the adiabatic theorem with explicit definitions of constants, and (b) we bound error in the adiabatic approxima-tion under conditions of noise and experimental error. We apply the new results to a standard example of violation of the adiabatic approximation, and to a supercon-ducting flux qubit. Further, adiabatic quantum computation requires large ground-state energy gaps throughout a Hamiltonian evolution if it is to solve problems in polynomial time. We identify a class of random Hamiltonians with non-nearest-neighbor interactions and a ground-state energy gap of O(1/√n), where n is the number of qubits. We also iden-tify two classes of Hamiltonians with non-nearest-neighbor interactions whose ground state can be found in polynomial time with adiabatic quantum computing. We then use the Jordan-Wigner transformation to derive equivalent results for Hamiltonians defined using Pauli operators.