Kernel Alignment for Quantum Support Vector Machines Using Genetic Algorithms

TL;DR

This paper introduces a genetic algorithm-based framework for optimizing quantum support vector machine kernels, improving classification accuracy and reducing manual design efforts across various datasets.

Contribution

It presents an automated method for designing quantum encoding circuits using genetic algorithms, outperforming standard techniques in QSVM applications.

Findings

GA-generated circuits match or surpass classical and quantum kernels

Positive correlation between test accuracy and quantum kernel entropy

Automated framework reduces trial and error in QSVM kernel design

Abstract

The data encoding circuits used in quantum support vector machine (QSVM) kernels play a crucial role in their classification accuracy. However, manually designing these circuits poses significant challenges in terms of time and performance. To address this, we leverage the GASP (Genetic Algorithm for State Preparation) framework for gate sequence selection in QSVM kernel circuits. We explore supervised and unsupervised kernel loss functions' impact on encoding circuit optimisation and evaluate them on diverse datasets for binary and multiple-class scenarios. Benchmarking against classical and quantum kernels reveals GA-generated circuits matching or surpassing standard techniques. We analyse the relationship between test accuracy and quantum kernel entropy, with results indicating a positive correlation. Our automated framework reduces trial and error, and enables improved QSVM based machine learning performance for finance, healthcare, and materials science applications.