Systems and methods for operating a quantum processor to determine energy eigenvalues of a hamiltonian
View Patent ↗Systems and methods for employing macroscopic resonant tunneling operations in quantum processors are described. New modes of use for quantum processor architectures employ probe qubits to determine energy eigenvalues of a problem Hamiltonian through macroscopic resonant tunneling operations. A dedicated probe qubit design that may be added to quantum processor architectures is also described. The dedicated probe qubit enables improved performance of macroscopic resonant tunneling operations and, consequently, improved performance of the new modes of use described.
1. A method comprising:
programming a quantum processor including a plurality of superconducting flux qubits with a problem Hamiltonian, wherein the problem Hamiltonian has at least one energy eigenvalue;
initializing a probe qubit from the plurality of superconducting flux qubits; and
scanning a transition rate of the probe qubit for a range of energy bias values of the probe qubit, wherein the transition rate of the probe qubit at least partially depends on a difference between an energy bias value of the probe qubit and an energy eigenvalue of the problem Hamiltonian such that the at least one energy eigenvalue of the problem Hamiltonian approximately corresponds to an energy bias value of the probe qubit to produce a locally maximal transition rate of the probe qubit, wherein programming the quantum processor comprises:
initializing a set of computation qubits from the plurality of superconducting flux qubits in the quantum processor with a respective energy bias for each qubit in the set of computation qubits and a respective tunneling energy for each qubit in the set of computation qubits; and
initializing communicative couplings between the computation qubits in the quantum processor with a respective coupling strength for each coupling device in the set of coupling devices.
2. The method of claim 1 wherein initializing a probe qubit from the plurality of superconducting flux qubits comprises:
programming a tunneling energy of the probe qubit, wherein the tunneling energy of the probe qubit is less than the respective tunneling energies of the computation qubits; and
communicatively coupling the probe qubit to at least one computation qubit.
3. The method of claim 2 wherein the probe qubit is a superconducting flux qubit from the plurality of superconducting flux qubits that is not included in the set of computation qubits.
4. The method of claim 2 , further comprising:
programming the at least one computation qubit to which the probe qubit is communicatively coupled with a compensation signal to compensate for the communicative coupling between the probe qubit and the at least one computation qubit.
5. The method of claim 4 wherein communicative coupling between the probe qubit and the at least one computation qubit is characterized by a coupling strength J, the method further comprising:
adding a compensation signal to the at least one computation qubit to which the probe qubit is communicatively coupled, wherein the compensation signal has a magnitude of at least approximately 2J and is of opposite sign to the coupling strength.
6. The method of claim 1 wherein scanning a transition rate of the probe qubit for a range of energy bias values of the probe qubit comprises, for a plurality of energy bias values for the probe qubit ranging from a first value to a second value, iteratively:
programming an energy bias of the probe qubit;
annealing the probe qubit; and
measuring the transition rate of the probe qubit.
7. The method of claim 6 wherein annealing the probe qubit includes lowering the tunneling energy of the probe qubit.
8. A method comprising:
programming a quantum processor including a plurality of superconducting flux qubits with a problem Hamiltonian, wherein the problem Hamiltonian has at least one energy eigenvalue;
initializing a probe qubit from the plurality of superconducting flux qubits; and
scanning a transition rate of the probe qubit for a range of energy bias values of the probe qubit, wherein the transition rate of the probe qubit at least partially depends on a difference between an energy bias value of the probe qubit and an energy eigenvalue of the problem Hamiltonian such that the at least one energy eigenvalue of the problem Hamiltonian approximately corresponds to an energy bias value of the probe qubit to produce a locally maximal transition rate of the probe qubit, wherein scanning a transition rate of the probe qubit for a range of energy bias values of the probe qubit comprises, for a plurality of energy bias values for the probe qubit ranging from a first value to a second value, iteratively:
programming an energy bias of the probe qubit;
annealing the probe qubit; and
measuring the transition rate of the probe qubit, wherein annealing the probe qubit includes lowering the tunneling energy of the probe qubit.