Quantum machine learning (QML) is widely proposed as a successor to classical models in cheminformatics, but rigorous benchmarks under realistic contemporary data constraints remain scarce. We compared two quantum classifiers (quantum support vector machine, variational quantum classifier) against gradient boosting, random forest, and neural network baselines across ten therapeutically diverse protein targets (45-18,286 compounds), using Bemis-Murcko scaffold-stratified cross-validation. Across 48 valid target-threshold tasks, no model class globally dominated (best-quantum versus best-classical Wilcoxon p = 0.189). Classical ensembles excelled on data-rich, structurally coherent targets (e.g., BACE1), while quantum kernels held a marginal edge on sparse, structurally bimodal targets (e.g., Mtb DprE1). A potential predictor of quantum advantage was the per-target best-quantum/best-classical model Cohen's κ gap (Spearman ρ = +0.685, p = 0.029), suggesting that decision calibration, not ranking, drives QML's edge. We propose the κ gap as a preliminary low-cost screening criterion for prioritizing QML in early stage discovery campaigns.
Quantum machine learning (QML) is widely proposed as a successor to classical models in cheminformatics, but rigorous benchmarks under realistic contemporary data constraints remain scarce. We compared two quantum classifiers (quantum support vector machine, variational quantum classifier) against gradient boosting, random forest, and neural network baselines across ten therapeutically diverse protein targets (45-18,286 compounds), using Bemis-Murcko scaffold-stratified cross-validation. Across 48 valid target-threshold tasks, no model class globally dominated (best-quantum versus best-classical Wilcoxon p = 0.189). Classical ensembles excelled on data-rich, structurally coherent targets (e.g., BACE1), while quantum kernels held a marginal edge on sparse, structurally bimodal targets (e.g., Mtb DprE1). A potential predictor of quantum advantage was the per-target best-quantum/best-classical model Cohen's κ gap (Spearman ρ = +0.685, p = 0.029), suggesting that decision calibration, not ranking, drives QML's edge. We propose the κ gap as a preliminary low-cost screening criterion for prioritizing QML in early stage discovery campaigns.
Quantum Machine Learning (QML) has emerged as an interdisciplinary field that harnesses the potential of quantum computing to revolutionize classical machine learning paradigms. This paper presents a benchmarking study focused on four quantum classifier algorithms: Quantum Support Vector Classifier (QSVC), Variational Quantum Circuit (VQC), Quantum Neural Network Classifier (QNNC) and Pegasos Quantum Support Vector Classifier (PegasosQSVC). These quantum classifiers hold promise for unlocking quantum advantages in classification tasks. However, a systematic assessment of their performance across diverse datasets is yet to be fully explored. In this research study, we evaluate and compare the classification performance of quantum machine learning classifiers on various datasets (Breast Cancer, Diabetes, and Titanic) and compare them to a classical model. Our findings revealed that PegasosQSVC consistently maintained high performance across all datasets with the highest average accuracy, recall, and F1-score values of 0.81, 0.72, and 0.71, respectively, followed by VQC, followed by VQC, while QSVC and QNNC perform variably. Nevertheless, QSVC was found to balance performance and training time, achieving an average accuracy of 0.81 and a training time of 177 seconds. Overall, the classical model surpassed QML models as it was up to 10% higher in average accuracy.
Abstract Over the past decade, quantum machine learning, particularly quantum support vector machines (QSVMs), has emerged as an optimistic alternative to classical machine learning (CML) techniques. This study rigorously benchmarks the performance of QSVM and CML-based models across four diverse datasets relevant to therapeutic proteins and peptides. Specifically, we evaluated these approaches for the prediction of B-cell epitopes (CLBtope), exosomal proteins (ExoPropred), hemolytic peptides (HemoPI), and toxic peptides (Toxinpred3). The maximum area under the receiver operating characteristic curve (AUC) for the CLBtope dataset achieved was 0.68 for QSVM and 0.82 for CML models. For the ExoPropred dataset, the maximum AUCs were 0.66 (QSVM) and 0.72 (CML). In contrast, both QSVM and CML models demonstrated high performance on the HemoPI dataset, yielding maximum AUCs of 0.95 and 0.98, respectively. Similarly, for the Toxinpred3 dataset, the maximum AUCs were 0.84 (QSVM) and 0.94 (CML). All models were evaluated using independent validation datasets not used during training. These results suggest that although CML currently demonstrates superior predictive capability for these tasks, the similar progression in performance indicates potential for future advancements in QSVM. Highlights Comparative study of QSVM and CML models on four bioinformatics datasets QSVM performance tries to approach CML in tasks involving hemolytic and toxic peptide prediction Independent validation confirms robustness of performance metrics Results highlight the potential of QSVMs as real-world quantum hardware continues to matures
Quantum Machine Learning (QML) has emerged as a paradigm that combines the computational advantages of quantum computing with the predictive capabilities of machine learning to address complex data classification problems. As data dimensionality increases rapidly and classical learning algorithms face scalability constraints, QML leverages quantum parallelism, entanglement, and high-dimensional Hilbert space representations to enhance learning performance. This paper reviews and analyses advanced QML algorithms, including Quantum Support Vector Machines (QSVM), Variational Quantum Circuits (VQC), Quantum Neural Networks (QNN), and Quantum Kernel Estimation, for optimizing both binary and multi-class classification tasks under high-complexity conditions. The proposed QML framework is evaluated on benchmark datasets like MNIST, the Breast Cancer Wisconsin (BCW) dataset, and synthetic nonlinear datasets, and is compared against classical machine learning baselines, including Support Vector Machines (SVM), Random Forests (RF), and Deep Neural Networks (DNN). The results demonstrate notable improvements in classification accuracy (up to 96.8%), decision margins, and computational efficiency in quantum-suitable data regimes, while also highlighting current limitations related to noise, circuit depth, and hardware constraints. Overall, the study presents a unified QML framework, theoretical formulations, and experimental evaluations that illustrate the potential of quantum algorithms for next-generation classification tasks.
Quantum machine learning (QML) benchmarks frequently compare quantum models against weak classical baselines -- the same methodological gap identified for quantum optimisation in the QCCB study (Popescu and Petrov, 2024). This paper presents the QML-DL Benchmark (QMLDB), a rigorous comparative evaluation of six QML model families against nine state-of-the-art classical deep learning baselines (MLP, CNN, LSTM, ResNet, Transformer, TabNet, XGBoost, LightGBM, CatBoost) across ten benchmark datasets spanning six task domains. QMLDB controls for model capacity, training budget, and data size. QML achieves top performance on three of ten datasets -- all three with strong quantum-process proximity. On the remaining seven, classical deep learning dominates, with Transformer and ResNet consistently outperforming all QML models by 4.8-18.4pp. QMLDB establishes that QML advantage is narrower than commonly claimed, confined to datasets with genuine quantum structure, and proposes a quantum-process proximity score (QPPS) to predict whether a new dataset will benefit from QML before hardware experiments.
Breast cancer continues to be among the most common causes of cancer death globally, with early and precise diagnosis playing a pivotal role in enhancing patient survival. Traditional machine learning (ML) techniques have shown great promise in the field of medical imaging and diagnosis; however, the emergence of quantum machine learning (QML) offers new avenues for the improvement of pattern discovery and diagnostic accuracy in high-dimensional medical data. This work describes a thorough benchmark comparison of quantum and classical machine learning methods for breast cancer diagnosis over the Wisconsin Diagnostic Breast Cancer data and mammographic image data. We compare variational quantum classifiers (VQC), quantum kernel methods (QKM), quantum convolutional neural networks (QCNNs), and hybrid quantum-classical neural structures with traditional classical baselines like support vector machines (SVM) and convolutional neural networks (CNN). QKM techniques perform better in high-dimensional feature spaces and better generalize on external validation sets. This implementation framework describes in depth how to develop, train, and deploy QML models in the clinical workflow, including optimization approaches, code structures, and deployment issues. The work demonstrates that it is possible to streamline breast cancer screening and diagnosis using QML with more accurate and more efficient solutions, which would lead to a huge impact on patient outcomes.
Quantum kernel methods have been proposed as a promising approach for leveraging near-term quantum computers for supervised learning, yet rigorous benchmarks against strong classical baselines remain scarce. We present a comprehensive empirical study of quantum kernel support vector machines (QSVMs) across nine binary classification datasets, four quantum feature maps, three classical kernels, and multiple noise models, totalling 970 experiments with strict nested cross-validation.
Our analysis spans four phases: (i) statistical significance testing, revealing that none of 29 pairwise quantum-classical comparisons reach significance at α=0.05; (ii) learning curve analysis over six training fractions, showing steeper quantum slopes on six of eight datasets that nonetheless fail to close the gap to the best classical baseline; (iii) hardware validation on IBM ibm_fez (Heron r2), demonstrating kernel fidelity r≥0.976 across six experiments; and (iv) seed sensitivity analysis confirming reproducibility (mean CV 1.4%).
A Kruskal-Wallis factorial analysis reveals that dataset choice dominates performance variance (ε2=0.73), while kernel type accounts for only 9%. Spectral analysis offers a mechanistic explanation: current quantum feature maps produce eigenspectra that are either too flat or too concentrated, missing the intermediate profile of the best classical kernel, the radial basis function (RBF). Quantum kernel training (QKT) via kernel-target alignment yields the single competitive result -- balanced accuracy 0.968 on breast cancer -- but with ~2,000x computational overhead.
Our findings provide actionable guidelines for quantum kernel research. The complete benchmark suite is publicly available to facilitate reproduction and extension.