Enhancing quantum support vector machines through variational kernel training

We introduce a new model in quantum machine learning (QML) that combines the strengths of existing quantum kernel SVM (QK-SVM) and quantum variational SVM (QV-SVM) methods. Our proposed model, quantum variational kernel SVM (QVK-SVM), utilizes quantum kernel and quantum variational algorithms to improve accuracy in QML applications. In this paper, we conduct extensive experiments on the Iris dataset to evaluate the performance of QVK-SVM against QK-SVM and QV-SVM models. Our results demonstrate that QVK-SVM outperforms both existing models regarding accuracy, loss, and confusion matrix indicators. We believe that QVK-SVM can be a reliable and transformative tool for QML applications and recommend its use in future QML research.