Abstract
Recent advancements in Quantum Computing and Machine Learning have increasedattention to Quantum Machine Learning (QML), which aims to develop machinelearning models by exploiting the quantum computing paradigm. One of the widelyused models in this area is the Variational Quantum Circuit (VQC), a hybridmodel where the quantum circuit handles data inference while classicaloptimization adjusts the parameters of the circuit. The quantum circuitconsists of an encoding layer, which loads data into the circuit, and atemplate circuit, known as the ansatz, responsible for processing the data.This work involves performing an analysis by considering both Amplitude- andAngle-encoding models, and examining how the type of rotational gate appliedaffects the classification performance of the model. This comparison is carriedout by training the different models on two datasets, Wine and Diabetes, andevaluating their performance. The study demonstrates that, under identicalmodel topologies, the difference in accuracy between the best and worst modelsranges from 10% to 30%, with differences reaching up to 41%. Moreover, theresults highlight how the choice of rotational gates used in encoding cansignificantly impact the model's classification performance. The findingsconfirm that the embedding represents a hyperparameter for VQC models.