Compositional optimization of quantum circuits for quantum kernels of support vector machines

TL;DR

This paper introduces a Bayesian algorithm that adaptively constructs quantum kernels for support vector machines, significantly improving classification performance over classical models with minimal training data.

Contribution

It presents a novel Bayesian approach to optimize quantum circuit structures for quantum kernels in SVMs, enhancing their classification capabilities.

Findings

Quantum models outperform classical models on tested classification tasks.

The algorithm efficiently increases quantum circuit complexity.

Quantum kernels require less training data for effective classification.

Abstract

While quantum machine learning (ML) has been proposed to be one of the most promising applications of quantum computing, how to build quantum ML models that outperform classical ML remains a major open question. Here, we demonstrate a Bayesian algorithm for constructing quantum kernels for support vector machines that adapts quantum gate sequences to data. The algorithm increases the complexity of quantum circuits incrementally by appending quantum gates selected with Bayesian information criterion as circuit selection metric and Bayesian optimization of the parameters of the locally optimal quantum circuits identified. The goal is to build quantum kernels for SVM that can solve classification problems with as little training data as possible. The performance of the resulting quantum models for the classification problems considered here significantly exceeds that of optimized classical models with conventional kernels.