Assessing Projected Quantum Kernels for the Classification of IoT Data

Abstract: The use of quantum computing for machine learning is among the most exciting applications of quantum technologies. Researchers are developing quantum models inspired by classical ones to find some possible quantum advantages over classical approaches. A major challenge in the development and testing of Quantum Machine Learning (QML) algorithms is the lack of datasets specifically designed for quantum algorithms. Existing datasets, often borrowed from classical machine learning, need modifications to be compatible with current noisy quantum hardware. In this work, we utilize a dataset generated by Internet-of-Things (IoT) devices in a format directly compatible with the proposed quantum algorithms, eliminating the need for feature reduction. Among quantum-inspired machine learning algorithms, the Projected Quantum Kernel (PQK) stands out for its elegant solution of projecting the data encoded in the Hilbert space into a classical space. We detail how a PQK approach can be employed to construct a prediction model on IoT data. We compare PQK with common Quantum Kernel methods and their classical counterparts, while also investigating the impact of various feature maps in encoding classical IoT data into quantum computers.