JP-7857499-B2 - Quantum computer device, method of operation, and non-temporary computer-readable storage medium

Inventors

• ドレイヤー ヘンリック
• イクバル モフシン

Assignees

• クオンティニュアム ゲーエムベーハー

Dates

Publication Date

20260512

Application Date

20230913

Priority Date

20220913

Claims (18)

1. A method for configuring a hybrid computer mechanism to perform chemical simulations, wherein the hybrid computer mechanism includes a combination of a classical computer coupled to a quantum computer, and the hybrid computer mechanism is configured to receive input data and generate corresponding processed output data from the input data when in use, and the method is (a) Configuring the classical computer to receive information describing the chemical system in the input data, (b) Configuring the classical computer to process the information describing the chemical system using a pre- entangler to generate a fixed circuit describing the static correlation of the wave functions describing the chemical system, and to generate a variational circuit describing the dynamic correlation of the wave functions using quantum Ansatz , (c) Configuring the quantum computer to execute quantum circuits corresponding to the fixed circuit and the variational circuit and generate quantum computation results, (d) Configuring the classical computer to process the quantum computation results and generate output data that includes information describing the electron orbital simulation of the chemical system, A method that includes this.
2. The method according to claim 1, comprising configuring the pre- entangler to function as a parameter-free pre-entangler.
3. The method according to claim 1 , comprising constructing the pre- entangler using a Matrix Product States (MPS) algorithm.
4. The method according to claim 3, comprising generating a matrix product state (MPS) based on a linear combination of unitary elements describing the chemical system.
5. The method according to claim 3, comprising configuring the hybrid computer mechanism to generate matrix product states (MPS) using a density matrix renormalization group (DMRG) algorithm in order to capture the complete active space (CAS) for one or more nonlinear transition metal complexes contained in the chemical system.
6. The method according to claim 3, comprising configuring the hybrid computer mechanism to generate matrix product states (MPS) by using an MPS algorithm based on sequential generation using ancilla qubits .
7. The method according to claim 1, comprising using the quantum circuit to find the ground state of the Hamiltonian of the chemical system .
8. The method according to claim 1, comprising configuring the variational circuit as a variational quantum eigenvalue solver based on one or more canonical transformations, wherein the method comprises configuring the hybrid computer mechanism to generate a density matrix renormalization group (DMRG) algorithm using the classical computer to construct static correlations in a wave function describing the chemical system, the density matrix renormalization group (DMRG) algorithm being used to generate the corresponding fixed portion of the quantum circuit.
9. (i) The operation of configuring the classical computer to process the information describing the chemical system, and assigning core space, virtual space, and molecular orbitals capable of occupying the active space to the active space based on the chemical system , (ii) The classical computer is configured to approximate the static correlation of the Hamiltonian of the chemical system by using the DMRG algorithm as the pre-entangler, and the operation of generating a matrix product state (MPS) description of the ground state for a given coupling dimension D |ψ 0 ||, (iii) An operation to generate a quantum circuit on the quantum computer that creates the MPS description |ψ 0 〉, (iv) The operation of constructing a variational quantum circuit that describes dynamic correlation by coupling orbitals in the core space , active space and virtual space , (v) An operation to minimize the ground state energy and generate an output result from the results of executing the quantum circuit, The method according to claim 1, comprising configuring the hybrid computer mechanism to perform the following.
10. A hybrid computer system configured to perform chemical simulations, wherein the hybrid computer system includes a combination of a classical computer coupled to a quantum computer, and the hybrid computer system is configured to receive input data and generate corresponding processed output data from the input data when in use, and the computer system is (a) Configured to receive information describing the chemical system in the input data, (b) The system is configured to process the information describing the chemical system using a pre- entangler to generate a fixed circuit describing the static correlation of the wave functions describing the chemical system, and to generate a variational circuit describing the dynamic correlation of the wave functions using quantum Ansatz , (c) A quantum circuit corresponding to the fixed circuit and the variational circuit is configured to execute and generate a quantum computation result, (d) A hybrid computer mechanism configured to process the quantum computation results and generate output data containing information describing the electron orbital simulation of the chemical system.
11. The hybrid computer mechanism according to claim 10, wherein the pre- entangler is configured to function as a parameter-free pre-entangler.
12. The hybrid computer mechanism according to claim 10, wherein the pre- entangler is configured to use a matrix multiplication state (MPS) algorithm.
13. The hybrid computer mechanism according to claim 12, wherein the hybrid computer mechanism is configured to generate matrix product states (MPS) using a density matrix renormalization group (DMRG) algorithm in order to capture the fully active space (CAS) for one or more nonlinear transition metal complexes contained in the chemical system.
14. The hybrid computer mechanism according to claim 10, wherein the hybrid computer mechanism is configured to generate matrix multiplication states (MPS) by using an MPS algorithm based on sequential generation using ancilla qubits .
15. The hybrid computer mechanism according to claim 10, wherein the hybrid computer mechanism is configured to use the quantum circuit to find the ground state of the Hamiltonian of the chemical system .
16. The hybrid computer mechanism according to claim 9, wherein the hybrid computer mechanism is configured to include the variational circuit as a variational quantum eigenvalue solver based on one or more canonical transformations, and the hybrid computer mechanism is configured to use the classical computer to generate a density matrix renormalization group (DMRG) algorithm to construct static correlations in wave functions describing the chemical system, and the density matrix renormalization group (DMRG) algorithm is used to generate the corresponding fixed portion of the quantum circuit.
17. The aforementioned hybrid computer mechanism, (i) Assigning core space, virtual space, and molecular orbitals capable of occupying the active space to the active space based on the chemical system , (ii) Approximating the ground state of the Hamiltonian by using the DMRG algorithm as the pre-entangler , and generating a matrix product state (MPS) description of the ground state |ψ 0 〉 for a given coupling dimension D, (iii) Finding a quantum circuit on the quantum computer that creates the MPS description |ψ 0 〉, (iv) Constructing a variational quantum circuit that describes dynamic correlation by coupling active orbitals in the core space and active space, (v) Based on the results of executing the quantum circuit, minimize the ground state energy to generate an output result, A hybrid computer mechanism according to claim 9, configured to perform the following:
18. A non-temporary computer-readable storage medium comprising a unique computer-readable instruction executable by data processing hardware, wherein the unique computer-readable instruction, when executed using the data processing hardware, implements the method according to any one of claims 1 to 9.

Description

This disclosure relates to a quantum computing device, such as a hybrid computing device including a combination of a quantum computer and a classical binary computer coupled to it. Furthermore, this disclosure relates to a method for operating such a quantum computing device. In addition, this disclosure relates to a software product recorded on a machine-readable medium, the software product being executable on the aforementioned quantum computing device for carrying out the aforementioned method. Quantum computers have recently become available as noisy intermediate-scale quantum (NISQ) devices [1], and furthermore, quantum computers can be classically simulated (i.e., "emulated"). One of the technical problems associated with NISQ devices is that their currently available qubits are too noisy to perform the deep quantum circuits required for most useful applications of quantum computers. For example, qubit phase estimation is difficult to implement on NISQ devices. To bridge the gap between modern NISQ devices and future fault-tolerant quantum computing devices, many researchers are now using variational quantum algorithms such as variational quantum eigensolvers (VQEs), quantum optimization algorithms, variational imaginary-time evolution algorithms, variational quantum adiabatic algorithms, and quantum neural network algorithms [2–8]. Variational quantum circuits executable using NISQ devices are typically shallower than many other types of quantum circuits executed using such devices. Due to this shallowness, variational quantum circuits exhibit a certain algorithmic resilience against noise, which is beneficial. Furthermore, such variational quantum circuits function iteratively by evaluating a given objective function of a given optimization problem on a given NISQ device, and updating the variational parameters of the variational quantum circuit using a classical optimization algorithm. A drawback is that designing a sufficiently expressive quantum circuit ansatz presents a problem of exponentially vanishing gradients as a function of the number of qubits used, resulting in a barren plateau that requires exponential time to escape [9]. In the presence of noise, the barren plateau becomes a problem for even less expressive, problem-inspired ansatz, as long as one attempts to train hyperlinear parameterized gates [10]. Therefore, reducing the number of variational parameters is essential for the success of variational quantum algorithms in the near future. Vanishing gradients also presented an early challenge to deep neural networks in classical machine learning. Generally, according to classical canonical transformation theory, the electron correlations of molecules are typically divided into two components. Static correlations have been used to quantify the portion of electron correlation associated with multiple relevant determinants whose energies are close to the Highest Occupied Molecular Orbital (HOMO) and Lowest Un-occupied Molecular Orbital (LUMO). Dynamic correlations describe the correction to the ground state resulting from excitations involving low-position core orbitals or high-position virtual orbitals. When static correlations are negligible, this is called single-reference chemistry. In such situations, the dynamic correlations can generally be adequately described by the classical coupled cluster (CC) method, provided that a sufficiently large ground system is used [19]. Conversely, multi-reference situations arise when a single determinant is insufficient to describe the chemical bonding, even qualitatively. In practice, these situations are observed in chemical reactions, specifically when crossing the HOMO-LUMO gap where the valence configurations of the products and reactants are almost entirely absent, as well as in excited states and transition metal chemistry [20]. To address the multi-reference situation, when performing quantum chemical simulations, the conventional method is to divide the molecular orbital into core (c) orbitals, active (a) orbitals, and virtual (v) orbitals, as shown in Figure 1b. (i) The core orbit is defined as being completely satisfied, (ii) The orbitals in the active space are partially filled, (iii) The orbit in virtual space is empty. The optimal method for assigning active spaces is unknown and initially relies on chemical intuition. Modern computer software programs such as BLOCK[21] and AutoCAS[22-25] use Fieldler vectors as a proxy for inter-orbital information and employ machine learning algorithms to automate the assignment of active spaces. And Hilbert spaces are Although it is demarcated as such, through this disclosure it is assumed that the Jordan-Wigner mapping from the fermion Fock space to the qubit to which the Slater determinant corresponds to the product state, and |1〉 (|0〉) represent occupied (unoccupied) orbits. To address multi-reference situations, many methods have been developed, such as multi-reference coupled clusters [26],