Quantum machine learning model feature space generation

Patent metadata

Field	Value
Publication number	US-12555046-B2
Application number	US-202117353268-A
Country	US
Kind code	B2
Filing date	Jun 21, 2021
Priority date	Jun 21, 2021
Publication date	Feb 17, 2026
Grant date	Feb 17, 2026

Abstract

Techniques regarding generating an ensemble of quantum kernel-based learners for one or more quantum machine learning models are provided. For example, one or more embodiments described herein can comprise a system, which can comprise a memory that can store computer executable components. The system can also comprise a processor, operably coupled to the memory, and that can execute the computer executable components stored in the memory. The computer executable components can comprise an ensemble component that can generate an ensemble of quantum kernel-based learners by selecting a quantum kernel at multiple iterations of a boosting procedure that analyzes a range of feature maps employable by a quantum machine learning model.

First claim

What is claimed is:

1 . A system, comprising:
a memory that stores computer executable components; and a processor, operably coupled to the memory, and that executes the computer executable components stored in the memory, wherein the computer executable components comprise:
an ensemble component that generates an ensemble of quantum kernel-based learners by selecting a quantum kernel at multiple iterations of a boosting procedure that analyzes a range of feature maps and Pauli rotation factors employable by a quantum machine learning model.

2 . The system of claim 1 , further comprising:
a training component that trains a set of quantum kernel-based learners on a training dataset for a plurality of quantum kernels, wherein the plurality of quantum kernels define permutations of the range of feature maps and Pauli rotation factors; and a test component that generates a performance metric associated with execution of the set of quantum kernel-based learners on a testing dataset.

3 . The system of claim 2 , wherein the test component generates the performance metric with each of the multiple iterations, wherein the ensemble component selects the quantum kernel based on the performance metric, and wherein the quantum kernel is associated with a quantum kernel-based learner comprised within the ensemble.

4 . The system of claim 2 , wherein a first quantum kernel is selected from the plurality of quantum kernels by the ensemble component at a first iteration of the boosting procedure, and wherein the system further comprises:
a modification component that generates a modified plurality of quantum kernels by removing the first quantum kernel from the plurality of quantum kernels based on the first iteration, wherein a second quantum kernel is selected from the modified plurality of quantum kernels at a second iteration of the boosting procedure.

5 . The system of claim 4 , further comprising:
an error weight component that adjusts a weight value associated with a sample of the training dataset based on an error performed by the at least one quantum kernel-based learner from the set of quantum kernel-based learners during the training performed by the training component; and a learner weight component that adjusts a learner weight associated with a quantum kernel-based learner from the set of quantum kernel-based learners based on the performance metric.

6 . The system of claim 5 , wherein error was performed with regards to the sample, and wherein the error weight component increases the weight value.

7 . The system of claim 5 , wherein the quantum kernel-based learner implemented the first quantum kernel during testing performed by the test component.

8 . The system of claim 5 , further comprising:
an ensemble component that generates the ensemble of quantum kernel-based learners by combining quantum kernel-based learners associated with the first quantum kernel and the second quantum kernel selected from the multiple iterations of the boosting procedure.

9 . The system of claim 8 , further comprising:
a model component that implements the quantum machine learning model with the ensemble of quantum kernel-based learners.

10 . The system of claim 1 , wherein the ensemble of quantum kernel-based learners comprises a plurality of quantum kernel-based learners executable on superconducting qubits.

11 . A computer-implemented method, comprising:
generating, by a system operatively coupled to a processor, an ensemble of quantum kernel-based learners by selecting a quantum kernel at multiple iterations of a boosting procedure that analyzes a range of feature maps and Pauli rotation factors employable by a quantum machine learning model.

12 . The computer-implemented method of claim 11 , further comprising:
training, by the system, a set of quantum kernel-based learners on a training dataset for a plurality of quantum kernels, wherein the plurality of quantum kernels define permutations of the range of feature maps and Pauli rotation factors; and generating, by the system, a performance metric associated with execution of the set of quantum kernel-based learners on a testing dataset.

13 . The computer-implemented method of claim 12 , wherein a first quantum kernel is selected from the plurality of quantum kernels at a first iteration of the boosting procedure based on the performance metric, wherein the quantum kernel is associated with a quantum kernel-based learner comprised within the ensemble, and wherein the computer-implemented method further comprises:
generating, by the system, a modified plurality of quantum kernels by removing the first quantum kernel from the plurality of quantum kernels based on the first iteration, wherein a second quantum kernel is selected from the modified plurality of quantum kernels at a second iteration of the boosting procedure.

14 . The computer-implemented method of claim 13 , further comprising:
adjusting, by the system, a weight value associated with a sample of the training dataset based on an error performed by the at least one quantum kernel-based learner from the set of quantum kernel-based learners during the training; and adjusting, by the system, a learner weight associated with a quantum kernel-based learner from the set of quantum kernel-based learners based on the performance metric.

15 . The computer-implemented method of claim 14 , further comprising:
generating, by the system, the ensemble of quantum kernel-based learners by combining quantum kernel-based learners associated with the first quantum kernel and the second quantum kernel selected from the multiple iterations of the boosting procedure.

16 . A computer program product for developing a feature space of a quantum machine learning model, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to:
generate, by the processor, an ensemble of quantum kernel-based learners by selecting a quantum kernel at multiple iterations of a boosting procedure that analyzes a range of feature maps and Pauli rotation factors employable by the quantum machine learning model.

17 . The computer program product of claim 16 , wherein the program instructions further cause the processor to:
train, by the processor, a set of quantum kernel-based learners on a training dataset for a plurality of quantum kernels, wherein the plurality of quantum kernels define permutations of the range of feature maps and Pauli rotation factors; and generate, by the processor, a performance metric associated with execution of the set of quantum kernel-based learners on a testing dataset.

18 . The computer program product of claim 17 , wherein a first quantum kernel is selected from the plurality of quantum kernels at a first iteration of the boosting procedure, based on the performance metric, wherein the quantum kernel is associated with a quantum kernel-based learner comprised within the ensemble, and wherein the program instructions further cause the processor to:
generate, by the processor, a modified plurality of quantum kernels by removing the first quantum kernel from the plurality of quantum kernels based on the first iteration, wherein a second quantum kernel is selected from the modified plurality of quantum kernels at a second iteration of the boosting procedure.

19 . The computer program product of claim 18 , wherein the program instructions further cause the processo

Assignees

• IBM

Inventors

• Rastunkov Vladimir
• Park Jae-Eun
• Mitra Abhijit

Classifications

• G06F18/2148 Primary

  characterised by the process organisation or structure, e.g. boosting cascade · CPC title

• G06N10/00

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

• G06F11/3428

  Benchmarking · CPC title

• G06N10/60

  Quantum algorithms, e.g. based on quantum optimisation, quantum Fourier or Hadamard transforms · CPC title

• G06N3/045

  Combinations of networks · CPC title