SYSTEMS AND METHODS FOR QUANTUM COMPUTING-ASSISTED PORTFOLIO SELECTION

Abstract

A method for quantum computing-assisted portfolio selection may include a classical computer program: (1) receiving a plurality of asset selection parameters for an asset portfolio; (2) initializing a current selection of assets from a plurality of available assets; (3) setting a risk upper bound value to a risk upper bound initial value, and a risk lower bound value to a risk lower bound initial value; (4) instructing a quantum computer to solve a first sub-problem; (5) calculating an objective functional value; (6) setting the risk upper bound value to the objective functional value; (7) instructing the quantum computer to determine a new selection of assets by solving a second sub-problem using a second quantum algorithm; and (8) returning an optimal portfolio selection.

Description

Claims

1. A method for quantum computing-assisted portfolio selection, comprising: receiving, by a classical computer program executed by a classical computer, a plurality of asset selection parameters for an asset portfolio;initializing, by the classical computer program, a current selection of assets from a plurality of available assets;setting, by the classical computer program, a risk upper bound value to a risk upper bound initial value, and a risk lower bound value to a risk lower bound initial value;instructing, by the classical computer program, a quantum computer to solve a first sub-problem;calculating, by the classical computer program, an objective functional value;setting, by the classical computer program, the risk upper bound value to the objective functional value;instructing, by the classical computer program, the quantum computer to determine a new selection of assets by solving a second sub-problem using a second quantum algorithm; andreturning, by the classical computer program, an optimal portfolio selection.

2. The method of claim 1, further comprising: determining an objective function that identifies a minimum amount of risk for the current selection of assets using a first quantum algorithm.

3. The method of claim 2, wherein calculating the objective functional value comprises using the objective function and a transaction cost for the current selection of assets.

4. The method of claim 1, further comprising: updating, by the classical computer program, the risk lower bound value based on the risk upper bound value and a difference between a transaction cost between the current selection of assets and the new selection of assets.

5. The method of claim 1, further comprising: creating, by the classical computer program, a constraint Hamiltonian by adding an integer cut to exclude a subset of the current selection of assets from the constraint Hamiltonian.

6. The method of claim 1, wherein the first quantum algorithm comprises the Harrow-Hassidim-Lloyd (HHL) quantum algorithm or the Variational Quantum Linear Solver (VQLS) quantum algorithm and the second quantum algorithm comprises binary optimization or the Quantum Approximate Optimization Algorithm (QAOA).

7. The method of claim 1, wherein the quantum computer comprises a universal quantum computer or quantum annealing hardware.

8. The method of claim 1, wherein the risk upper bound initial value is set to infinity, and a risk lower bound initial value is set to negative infinity.

9. The method of claim 1, wherein the quantum computer solves the second sub-problem by minimizing a constraint Hamiltonian.

10. A system, comprising: a classical computer comprising a memory storing a classical computer program and a computer processor; anda quantum computer in communication with the classical computer;wherein the classical computer program receives a plurality of asset selection parameters for an asset portfolio; initializes a current selection of assets from a plurality of available assets; sets a risk upper bound value to a risk upper bound initial value, and a risk lower bound value to a risk lower bound initial value; instructs a quantum computer to solve a first sub-problem; calculates an objective functional value; sets the risk upper bound value to the objective functional value; instructs the quantum computer to determine a new selection of assets by solving a second sub-problem using a second quantum algorithm; and returns an optimal portfolio selection.

11. The system of claim 10, wherein the classical computer program determines an objective function that identifies a minimum amount of risk for the current selection of assets using a first quantum algorithm.

12. The system of claim 10, wherein the classical computer program calculates the objective functional value, the objective function, and a transaction cost for the current selection of assets.

13. The system of claim 10, wherein the classical computer program updates the risk lower bound value based on the risk upper bound value and a difference between a transaction cost between the current selection of assets and the new selection of assets.

14. The system of claim 10, wherein the classical computer program creates a constraint Hamiltonian by adding an integer cut to exclude a subset of the current selection of assets from the constraint Hamiltonian.

15. The system of claim 10, wherein the quantum computer solves the first sub-problem using the Harrow-Hassidim-Lloyd (HHL) quantum algorithm or the Variational Quantum Linear Solver (VQLS) quantum algorithm, and the second quantum algorithm comprises binary optimization or the Quantum Approximate Optimization Algorithm (QAOA).

16. The system of claim 11, wherein the classical computer program selects a quantum computer to execute the first quantum algorithm, wherein the quantum computer comprises a universal quantum computer or quantum annealing hardware.

17. The system of claim 11, wherein the risk upper bound initial value is set to infinity, and a risk lower bound initial value is set to negative infinity.

18. The system of claim 11, wherein the quantum computer solves the second sub-problem by minimizing a constraint Hamiltonian.

19. An electronic device, comprising: a memory storing a classical computer program; anda computer processor;wherein, when executed by the computer processor, the classical computer program causes the computer processor to: receive a plurality of asset selection parameters for an asset portfolio;initialize a current selection of assets from a plurality of available assets;set a risk upper bound value to a risk upper bound initial value, anda risk lower bound value to a risk lower bound initial value; instruct a quantum computer to solve a first sub-problem;calculate an objective functional value;set the risk upper bound value to the objective functional value;instruct the quantum computer to determine a new selection of assets by solving a second sub-problem using a second quantum algorithm; andreturn an optimal portfolio selection.

20. The electronic device of claim 19, wherein the classical computer program selects a quantum computer to solve the first sub-problem using a first quantum algorithm, wherein the quantum computer comprises a universal quantum computer or quantum annealing hardware, and the first quantum algorithm comprises the Harrow-Hassidim-Lloyd (HHL) quantum algorithm or the Variational Quantum Linear Solver (VQLS) quantum algorithm, and the second quantum algorithm comprises binary optimization or the Quantum Approximate Optimization Algorithm (QAOA).