Hybrid quantum computing system for hyper parameter optimization in machine learning

1 . A computer system for performing optimization of hyper parameters in machine learning, comprising:

a classical computer apparatus comprising:

a processor;

a memory; and

an optimization application that is stored in the memory and executable by the processor;

a quantum optimizer in communication with the classical computer apparatus, the quantum optimizer comprising:

a quantum processor; and

a quantum memory;

wherein the optimization application is configured for:

gathering one or more data sets associated with an application;

identifying one or more parameters associated with the application, wherein the one or more parameters are hyperparameters that control machine learning models, wherein the hyperparameters comprise model architecture, learning rate, number of epochs, number of branches in a decision tree, and number of clusters in a clustering algorithm;

automatically select a machine learning model type for constructing a machine learning model;

constructing the machine learning model based on the machine learning model type using a first part of the one or more data sets and the one or more parameters, wherein the one or more parameters have an impact on an optimized version of the machine learning model;

determining one or more conditions associated with optimizing the machine learning model;

transmitting the machine learning model, the one or more data sets, and the one or more conditions to the quantum optimizer;

wherein the quantum optimizer is configured for:

in response to receiving the machine learning model, the one or more data sets, and the one or more conditions, retrieving a second part of the one or more data sets;

computing a set of optimal hyperparameters for the machine learning model based on the second part of the one or more data sets and the one or more conditions; and

transmitting the set of optimal hyperparameters to the optimization application; and

wherein the optimization application is configured for:

utilizing the set of optimal hyperparameters for the machine learning model to create the optimized version of the machine learning model for solving a problem associated with the application, via the machine learning model.

2 . The computer system of claim 1 , wherein the quantum optimizer is configured to computer the set of optimal hyperparameters, via quantum annealing method.

3 . The computer system of claim 1 , wherein the one or more conditions comprise at least one of minimizing a predefined function and maximizing the predefined function.

4 . The computer system of claim 1 , wherein the one or more data sets may comprise at least one of numerical data, numerical data, categorical data, time series data, and text data.

5 . The computer system of claim 1 , wherein the one or more data sets comprises the first part of the one or more data sets and the second part of the one or more data sets.

6 . The computer system of claim 1 , wherein the first part of the one or more data sets comprises the second part of the one or more data sets.

7 . A computer program product for performing optimization of hyper parameters in machine learning, comprising a non-transitory computer-readable storage medium having computer-executable instructions for causing a classical computer apparatus comprising an optimization application to:

gather one or more data sets associated with an application;

identify one or more parameters associated with the application, wherein the one or more parameters are hyperparameters that control machine learning models, wherein the hyperparameters comprise model architecture, learning rate, number of epochs, number of branches in a decision tree, and number of clusters in a clustering algorithm;

automatically select a machine learning model type for constructing a machine learning model;

construct the machine learning model based on the machine learning model type using a first part of the one or more data sets and the one or more parameters, wherein the one or more parameters have an impact on an optimized version of the machine learning model;

determine one or more conditions associated with optimizing the machine learning model;

transmit the machine learning model, the one or more data sets, and the one or more conditions to a quantum optimizer;

wherein the quantum optimizer is configured for:

in response to receiving the machine learning model, the one or more data sets, and the one or more conditions, retrieving a second part of the one or more data sets;

computing a set of optimal hyperparameters for the machine learning model based on the second part of the one or more data sets and the one or more conditions; and

transmitting the set of optimal hyperparameters to the optimization application; and wherein the optimization application is configured for:

utilizing the set of optimal hyperparameters for the machine learning model to create the optimized version of the machine learning model for solving a problem associated with the application, via the machine learning model.

8 . The computer program product of claim 7 , wherein the quantum optimizer is configured to compute the set of optimal hyperparameters, via quantum annealing method.

9 . The computer program product of claim 7 , wherein the one or more conditions comprise at least one of minimizing a predefined function and maximizing the predefined function.

10 . The computer program product of claim 7 , wherein the one or more data sets may comprise at least one of numerical data, numerical data, categorical data, time series data, and text data.

11 . The computer program product of claim 7 , wherein the one or more data sets comprises the first part of the one or more data sets and the second part of the one or more data sets.

12 . A method for performing optimization of hyper parameters in machine learning, comprising:

gathering, via a classical computer apparatus, one or more data sets associated with an application;

identifying, via the classical computer apparatus, one or more parameters associated with the application, wherein the one or more parameters are hyperparameters that control machine learning models, wherein the hyperparameters comprise model architecture, learning rate, number of epochs, number of branches in a decision tree, and number of clusters in a clustering algorithm;

automatically selecting a machine learning model type for constructing a machine learning model;

constructing, via the classical computer apparatus, the machine learning model based on the machine learning model type using a first part of the one or more data sets and the one or more parameters, wherein the one or more parameters have an impact on an optimized version of the machine learning model;

determining, via the classical computer apparatus, one or more conditions associated with optimizing the machine learning model;

transmitting, via the classical computer apparatus, the machine learning model, the one or more data sets, and the one or more conditions to a quantum optimizer;

in response to receiving the machine learning model, the one or more data sets, and the one or more conditions, retrieving, via the quantum optimizer, a second part of the one or more data sets;

computing, via the quantum optimizer, a set of optimal hyperparameters for the machine learning model based on the second part of the one or more data sets and the one or more conditions;

transmitting, via the quantum optimizer, the set of optimal hyperparameters to the optimization application; and

utilizing, via the classical computer apparatus, the set of optimal hyperparameters for the machine learning model to create the optimized version of the machine learning model for solving a problem associated with the application, via the machine learning model.

13 . The method of claim 12 , wherein quantum optimizer is configured to compute the set of optimal hyperparameters, via quantum annealing method.

14 . The method of claim 12 , wherein the one or more conditions comprise at least one of minimizing a predefined function and maximizing the predefined function.

15 . The method of claim 12 , wherein the one or more data sets may comprise at least one of numerical data, numerical data, categorical data, time series data, and text data.

16 . The method of claim 12 , wherein the one or more data sets comprises the first part of the one or more data sets and the second part of the one or more data sets.

17 . The method of claim 12 , wherein the first part of the one or more data sets comprises the second part of the one or more data sets.