Variational Circuit Based Hybrid Quantum-Classical Algorithm VC-HQCA

Quantum machine learning (QML) has emerged as a promising field that combines principles from quantum computing and machine learning. This work investigates the influence of variational circuit design on the performance of QML hybrid models. A novel classification system is introduced to categorize variational circuit designs based on their architectural properties. Specifically, the research focuses on exploring the “ONE-TO-ONE-VC” and analyzes the diverse effects of different combinations of quantum gates on the circuit’s performance. Moreover, the research delves into the study of hyper-parameters and their impact on the performance of QML models. By systematically varying hyper-parameters, the objective is to understand their influence on the overall performance and efficiency of the models. Additionally, a hybrid quantum-classical classifier is built and bench-marked against current classifiers implemented in Qiskit, a widely-used quantum computing framework. The findings of this research provide compelling evidence for the significance of variational circuit design in QML models. By demonstrating the impact of circuit design on model performance.