An end-to-end trainable hybrid classical-quantum classifier

Chen, Samuel Yen-Chi; Huang, Chih-Min; Hsing, Chia-Wei; Kao, Ying-Jer

doi:10.48550/arxiv.2102.02416

Cited by 2 publications

(3 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Currently, one of the challenges of QML is the limited number of qubits, which restricts the use of quantum devices in applications with QML algorithms [2,23]. Therefore, using a quantum computer with complex large dimensionality data is still difficult.…”

Section: Principal Component Analysismentioning

confidence: 99%

An Efficient Binary Harris Hawks Optimization based on Quantum SVM for Cancer Classification Tasks

Houssein,

Abohashima,

Elhoseny

et al. 2022

Preprint

View full text Add to dashboard Cite

Cancer classification based on gene expression increases early diagnosis and recovery, but high-dimensional genes with a small number of samples are considered the major challenge in cancer classification tasks. Gene selection is a critical step in large-dimensional features to select the most informative and relevant subset of genes. The Gene selection problem utilizes metaheuristic techniques to extract the optimal subset of genes in microarray cancer datasets, such as particle swarm optimization (PSO), Harris hawks optimization (HHO), and grey wolf optimizer (GWO). Binary Harris hawk optimization (BHHO) technique, which mimics the behavior of the cooperative action of Harris hawks in nature, is recently proposed as one of these techniques. The biggest challenge of quantum hardware is the limited number of qubits, which restricts the use of quantum devices in real applications. The principal component analysis (PCA) is applied to reduce the selected genes to match the qubit numbers. In this work, a new hybrid quantum-kernel support vector machine (QKSVM) with BHHO called BHHO-PCA-QKSVM for cancer classification on a quantum simulator. This study aims to improve the microarray cancer prediction performance with the quantum kernel estimation based on the informative genes by BHHO. The quantum computer is used for estimation the kernel with the training data of the reduced genes and generation of the quantum kernel matrix. Besides, the classical computer is used for drawing the support vectors based on the quantum kernel matrix and applying the prediction stage. The colon and breast microarray datasets are used for evaluating the proposed approach performance with all genes and the selected genes. The proposed model enhances the overall performance of the two datasets. Also, the final results of the proposed model compared with different quantum feature maps (kernels) and classical radial basis function (RBF) kernel.

show abstract

Section: Principal Component Analysismentioning

confidence: 99%

An Efficient Binary Harris Hawks Optimization based on Quantum SVM for Cancer Classification Tasks

Houssein,

Abohashima,

Elhoseny

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…The corresponding expressions are too bulky to be presented explicitly, and possibly the best course of action is numerical calculation of coefficients from Eq. (12).…”

Section: Other Two-qubit Gatesmentioning

confidence: 99%

“…The emergent field of quantum machine learning (QML) gained significant attention in the last several years [6]. From the application perspective, QML algorithms for classification [7][8][9][10][11][12], generative modelling [13][14][15][16][17][18][19], reinforcement learning [20][21][22], and solving differential equations [23][24][25][26] were proposed recently. From the operational perspective, the algorithmic workflow has changed from a fault tolerance-oriented approach with deep ancilla-based circuits [6] to a hybrid quantum-classical approach [27,28].…”

Section: Introductionmentioning

confidence: 99%

Generalized quantum circuit differentiation rules

Kyriienko,

Elfving

2021

Preprint

View full text Add to dashboard Cite

Variational quantum algorithms that are used for quantum machine learning rely on the ability to automatically differentiate parametrized quantum circuits with respect to underlying parameters. Here, we propose the rules for differentiating quantum circuits (unitaries) with arbitrary generators. Unlike the standard parameter shift rule valid for unitaries generated by operators with spectra limited to at most two unique eigenvalues (represented by involutory and idempotent operators), our approach also works for generators with a generic non-degenerate spectrum. Based on a spectral decomposition, we derive a simple recipe that allows explicit derivative evaluation. The derivative corresponds to the weighted sum of measured expectations for circuits with shifted parameters. The number of function evaluations is equal to the number of unique positive nonzero spectral gaps (eigenvalue differences) for the generator. We apply the approach to relevant examples of two-qubit gates, among others showing that the fSim gate can be differentiated using four measurements. Additionally, we present generalized differentiation rules for the case of Pauli string generators, based on distinct shifts (here named as the triangulation approach), and analyse the variance for derivative measurements in different scenarios. Our work offers a toolbox for the efficient hardware-oriented differentiation needed for circuit optimization and operator-based derivative representation.

show abstract

An end-to-end trainable hybrid classical-quantum classifier

Cited by 2 publications

References 0 publications

An Efficient Binary Harris Hawks Optimization based on Quantum SVM for Cancer Classification Tasks

An Efficient Binary Harris Hawks Optimization based on Quantum SVM for Cancer Classification Tasks

Generalized quantum circuit differentiation rules

Contact Info

Product

Resources

About