Uncertainty aware parameter provision for a variational quantum algorithm

Patent metadata

Field	Value
Publication number	US-12468977-B2
Application number	US-202117378437-A
Country	US
Kind code	B2
Filing date	Jul 16, 2021
Priority date	Jul 16, 2021
Publication date	Nov 11, 2025
Grant date	Nov 11, 2025

Abstract

Systems, computer-implemented methods and/or computer program products that can facilitate providing a defined parameter, determining whether to employ the defined parameter for a variational quantum algorithm, and running the variational quantum algorithm on a quantum system, are provided. According to an embodiment, a system can comprise a memory that stores computer executable components and a processor that executes the computer executable components stored in the memory. The computer executable components can comprise a decision component that determines, based upon an uncertainty prediction regarding the usability of the defined parameter that has been output from a machine learning model, whether to employ the defined parameter in a variational quantum algorithm, such as run on a quantum system.

First claim

What is claimed is:

1 . A system, comprising:
a memory that stores computer executable components; and a processor that executes the computer executable components stored in the memory, wherein the computer executable components comprise:
a decision component that determines, based upon an uncertainty prediction regarding usability of a defined parameter that has been output from a machine learning model, whether to employ the defined parameter for running a variational quantum algorithm;
a training component that trains the machine learning model by employing a central data store having data related to the variational quantum algorithm, wherein the training comprises training a machine learning model to be uncertainty aware by basing the machine learning model on a natively uncertainty aware machine learning algorithm; and
an Ansatz component comprises a set of quantum circuits with one or more free parameters, and approximates a quantum state of interest in which the one or more free parameters take optimal values, wherein the decision component outputs feedback to the Ansatz component to direct the Ansatz component to perform parameter optimization on the one or more free parameters using the defined parameter.

2 . The system of claim 1 , further comprising:
a performance component that executes the machine learning model to provide the uncertainty prediction and the defined parameter, wherein the defined parameter is a variational parameter for initialization of the variational quantum algorithm.

3 . The system of claim 1 , further comprising:
an updating component that updates the central data store with the defined parameter and the associated uncertainty prediction.

4 . The system of claim 1 , further comprising:
an aggregation component that enables updating of the central data store with one or more other defined parameters, other associated uncertainty predictions, or a combination thereof, from a plurality of systems being distributed relative to one another.

5 . The system of claim 1 , further comprising:
a quantum calculation component that executes the variational quantum algorithm on a quantum device, wherein the variational quantum algorithm employs one or more parameters determined at least in part based on the determination regarding the defined parameter.

6 . The system of claim 1 , wherein the Ansatz component employs an Ansatz method to optimize a supplementary parameter where the decision component determines that the defined parameter will not be employed by the variational quantum algorithm.

7 . A computer-implemented method, comprising:
determining, by a decision component of a system operatively coupled to a processor, and based upon an uncertainty prediction regarding usability of a defined parameter having been output from a machine learning model, whether to employ the defined parameter for running a variational quantum algorithm; training, by the system, the machine learning model by employing, by the system, a central data store having data related to the variational quantum algorithm, wherein the training comprises training a machine learning model to be uncertainty aware by basing the machine learning model on a natively uncertainty aware machine learning algorithm; and approximating, by the system, a quantum state of interest in which one or more free parameters take optimal values, wherein the decision component outputs feedback to an Ansatz component to direct the Ansatz component to perform parameter optimization on the one or more free parameters associated with a set of quantum circuits in the Ansatz component using the defined parameter.

8 . The computer-implemented method of claim 7 , further comprising:
executing, by the system, the machine learning model to provide the uncertainty prediction and the defined parameter, wherein the defined parameter is a variational parameter for initialization of the variational quantum algorithm.

9 . The computer-implemented method of claim 7 , further comprising:
updating, by the system, the central data store with the defined parameter and the associated uncertainty prediction.

10 . The computer-implemented method of claim 7 , further comprising:
enabling, by the system, updating of the central data store with one or more other defined parameters, other associated uncertainty predictions, or a combination thereof, from a plurality of systems being distributed relative to one another.

11 . The computer-implemented method of claim 7 , further comprising:
executing, by the system, the variational quantum algorithm on a quantum device, including employing, by the system, one or more parameters, by the variational quantum algorithm, determined at least in part based on the determination regarding the defined parameter.

12 . The computer-implemented method of claim 7 , further comprising:
employing, by the system, an Ansatz method to optimize a supplementary parameter where the decision component determines that the defined parameter will not be employed by the variational quantum algorithm.

13 . The computer-implemented method of claim 7 , wherein the training the machine learning model to be uncertainty aware is by basing the machine learning model on a natively uncertainty aware machine learning algorithm such as a Gaussian process or a Bayesian neural network.

14 . A computer program product facilitating a process providing a defined parameter and determining whether to employ the defined parameter for a variational quantum algorithm, the computer program product comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to:
determine, by the processor, an uncertainty prediction, related to a defined parameter and relative to the variational quantum algorithm; determine, by a decision component in the processor, and based upon the uncertainty prediction regarding usability of the defined parameter having been output from a machine learning model, whether to employ the defined parameter for running a variational quantum algorithm; and approximate, by the processor, a quantum state of interest in which one or more free parameters take optimal values, wherein the decision component outputs feedback to an Ansatz component to direct the Ansatz component to perform parameter optimization on the one or more free parameters associated with a set of quantum circuits in the Ansatz component using the defined parameter.

15 . The computer program product of claim 14 , further comprising causing the processor to:
execute, by the processor, the machine learning model to provide the uncertainty prediction and the defined parameter, wherein the defined parameter is a variational parameter for initialization of the variational quantum algorithm.

16 . The computer program product of claim 14 , further comprising causing the processor to:
train, by the processor, the machine learning model by employing, by the system, a central data store having data related to the variational quantum algorithm.

17 . The computer program product of claim 16 , further comprising causing the processor to:
update, by the processor, the central data store with the defined parameter and the associated uncertainty prediction.

18 . The computer program product of claim 16 , further comprising causing the processor to:
enable, by the processor, updating of the central data store with one or more other defined parameters, other

Assignees

• IBM

Inventors

• Pyzer-Knapp Edward Oliver
• Motta Mario
• Johnston Michael

Classifications

• G06N10/00

  Quantum computing, i.e. information processing based on quantum-mechanical phenomena · CPC title

• G06N3/0475

  Generative networks · CPC title

• G06N3/047

  Probabilistic or stochastic networks · CPC title

• G06N20/10

  using kernel methods, e.g. support vector machines [SVM] · CPC title

• G06N20/00 Primary

  Machine learning · CPC title