Feature selection in high-dimensional data spaces continues to be an intrinsic challenge in machine learning algorithms, especially when the number of features regularly exceeds thousands. Traditional techniques including mutual information filters, LASSO regularization, and recursive feature elimination are limited by the computational inefficiency of brute force searching as well as poor scalability in the ultra-high-dimensional setting. In this paper, a novel technique, the Hybrid Quantum-Classical Variational Algorithm (HQCVA), that combines quantum computing and classical optimization methods in the Variational Quantum Eigensolver (VQE) approach is introduced. Specifically, the quantum part is responsible for representing the correlation structure of the input features via the Ising model. To reduce composite loss, the classical sub-system uses the Adam stochastic optimizer to update variational parameters iteratively while also balancing density between predictive accuracies, densifications and depth of quantum circuits. Using five benchmark datasets (22,000 gene data; 12,000 pixels of hyperspectral imagery; 8,500 data points of financial time series), HQCVA produces a 12-18% higher F1-score than leading, state-of-the-art classical systems while decreasing the number of selected features by 40-65%. All simulations of circuits are performed on IBM Qiskit Aer with noise added, to simulate Eagle r3 consoles. Furthermore, the results indicate that the quantum advantage scales with increasing feature dimension through this process to demonstrate that near-term quantum devices have practical applications in real-world feature selection applications.
Feature selection in high-dimensional data spaces continues to be an intrinsic challenge in machine learning algorithms, especially when the number of features regularly exceeds thousands. Traditional techniques including mutual information filters, LASSO regularization, and recursive feature elimination are limited by the computational inefficiency of brute force searching as well as poor scalability in the ultra-high-dimensional setting. In this paper, a novel technique, the Hybrid Quantum-Classical Variational Algorithm (HQCVA), that combines quantum computing and classical optimization methods in the Variational Quantum Eigensolver (VQE) approach is introduced. Specifically, the quantum part is responsible for representing the correlation structure of the input features via the Ising model. To reduce composite loss, the classical sub-system uses the Adam stochastic optimizer to update variational parameters iteratively while also balancing density between predictive accuracies, densifications and depth of quantum circuits. Using five benchmark datasets (22,000 gene data; 12,000 pixels of hyperspectral imagery; 8,500 data points of financial time series), HQCVA produces a 12-18% higher F1-score than leading, state-of-the-art classical systems while decreasing the number of selected features by 40-65%. All simulations of circuits are performed on IBM Qiskit Aer with noise added, to simulate Eagle r3 consoles. Furthermore, the results indicate that the quantum advantage scales with increasing feature dimension through this process to demonstrate that near-term quantum devices have practical applications in real-world feature selection applications.
Abstract Unsupervised feature selection is essential for high-dimensional machine learning tasks, as it improves model quality and efficiency while providing interpretable insights into datasets. However, existing methods often struggle to simultaneously achieve robustness, reliability, interpretability, and computational efficiency. Furthermore, feature selection in quantum machine learning remains largely unexplored despite the potential of quantum approaches to capture complex feature interactions. In this work, we propose the Quantum Hybrid Feature Selector, a novel unsupervised feature selection framework that combines quantum-enhanced feature extraction with interpretable scoring algorithms. Our main contributions are threefold: we introduce a quantum hybrid autoencoder pipeline that couples feature extraction with feature ranking in the original input space, we define three scoring mechanisms: SHAP-based scoring, correlation-matrix scoring, and weight-based analysis, that relate latent features to original features without requiring external predictive models, and we provide comprehensive empirical evaluation on synthetic and real-world benchmarks with statistical validation. On Madelon-style synthetic datasets, our quantum SCM achieves up to 23.5% improvement in Mean Informative Rank and 9.7% improvement in Informative Ratio compared to classical alternatives, with all improvements statistically significant at p<0.001 p < 0.001 . On the Communities and Crime dataset with injected noise features, both quantum and classical SCM variants achieve near-perfect noise elimination, demonstrating effective denoising capability on real-world data. We also show that expert input is essential for evaluating feature selectors, as different methods emphasize distinct aspects of the data despite similar aggregate metrics. Finally, we evaluate QHFS under realistic hardware noise conditions using device-informed simulations based on IBM superconducting quantum processors, demonstrating graceful performance degradation and systematic improvement with increased shot budgets, which supports the practical viability of our approach on near-term quantum devices.
Hybrid Quantum Classical (HQC) algorithms constitute one of the most effective paradigms for exploiting the computational advantages of quantum systems in large-scale numerical tasks. By operating in high-dimensional Hilbert spaces, quantum circuits enable exponential speed-ups and provide access to richer representations of cost landscapes compared to purely classical methods. These capabilities are particularly relevant for machine learning, where state-of-the-art models especially in Natural Language Processing (NLP) suffer from prohibitive time complexity due to massive matrix multiplications and high-dimensional optimization.
In this manuscript, we propose a Hybrid Quantum Classical selection mechanism for the Mamba architecture, designed specifically for temporal sequence classification problems. Our approach leverages Variational Quantum Circuits (VQCs) as quantum gating modules that both enhance feature extraction and improve suppression of irrelevant information. This integration directly addresses the computational bottlenecks of deep learning architectures by exploiting quantum resources for more efficient representation learning.
We analyze how introducing quantum subroutines into large language models (LLMs) impacts their generalization capability, expressivity, and parameter efficiency. The results highlight the potential of quantum-enhanced gating mechanisms as a path toward scalable, resource-efficient NLP models, in a limited simulation step. Within the first four epochs on a reshaped MNIST dataset with input format (batch, 784, d_model), our hybrid model achieved 24.6% accuracy while using one quantum layer and achieve higher expressivity, compared to 21.6% obtained by a purely classical selection mechanism. we state No founding
Hybrid quantum classical algorithms offer a promising approach for exploiting Noisy Intermediate Scale Quantum devices; however, their practical deployment remains limited by excessive circuit depth, hardware noise, and inefficient utilization of quantum resources. This paper presents a Hybrid Variational Optimization (HVO) framework that reduces quantum algorithmic complexity by integrating adaptive circuit pruning with classical feedback-driven optimization and a unified Quantum Complexity Metric (QCM). Unlike prior adaptive variational approaches such as RE-ADAPT-VQA, the proposed framework explicitly optimizes the trade-off between gate count, fidelity, and qubit resources while remaining modular and hardware-agnostic. To ensure a fair and transparent evaluation, the proposed method is compared against a well-tuned shallow Variational Quantum Eigensolver (VQE), QAOA, and RE-ADAPT-VQA. Experimental validation on IBM Quantum simulators and real NISQ hardware demonstrates up to a 35% reduction in circuit complexity, a 30% reduction in execution time, and improved fidelity of 0.94. Furthermore, performance evaluation using accuracy, precision, recall, and F1-score confirms the robustness and reliability of the proposed framework. These results demonstrate that HVO provides a scalable and noise-resilient solution for near term quantum computing applications.
Choosing which features to retain from a high-dimensional dataset is one of the more practically consequential decisions in building a machine learning pipeline, yet it is rarely treated as anything other than a preprocessing afterthought. In this paper we ask whether quantum computation can play a meaningful role in that decision. We present Variational Quantum Feature Selection (VQFS), a method that assigns trainable scalar weights to input features and folds those weights directly into the rotation angles of a parameterized quantum circuit. An Ll penalty on the weight vector encourages many weights to collapse toward zero during training, leaving a compact subset of features that the circuit has learned to rely on. Because the weights are differentiable, the whole system-feature selector and quantum classifier together-is trained end-to-end with gradient descent rather than through the combinatorial search that makes classical wrapper methods expensive. We built VQFS on top of the PennyLane simulator so that it runs on an ordinary laptop without access to quantum hardware. Testing on the Breast Cancer Wisconsin, Wine, and Iris benchmarks, we found that VQFS matched or beat five classical baselines (Lasso, PCA, Random Forest importance, Mutual Information, and RFE) on accuracy while cutting the feature count by 25-60%. Five-fold cross-validation showed the gains were consistent rather than the result of a lucky split. We see this work as a small but concrete step toward integrating quantum methods into the full machine learning workflow, not just the classification stage.
Choosing which features to retain from a high-dimensional dataset is one of the more practically consequential decisions in building a machine learning pipeline, yet it is rarely treated as anything other than a preprocessing afterthought. In this paper we ask whether quantum computation can play a meaningful role in that decision. We present Variational Quantum Feature Selection (VQFS), a method that assigns trainable scalar weights to input features and folds those weights directly into the rotation angles of a parameterized quantum circuit. An Ll penalty on the weight vector encourages many weights to collapse toward zero during training, leaving a compact subset of features that the circuit has learned to rely on. Because the weights are differentiable, the whole system-feature selector and quantum classifier together-is trained end-to-end with gradient descent rather than through the combinatorial search that makes classical wrapper methods expensive. We built VQFS on top of the PennyLane simulator so that it runs on an ordinary laptop without access to quantum hardware. Testing on the Breast Cancer Wisconsin, Wine, and Iris benchmarks, we found that VQFS matched or beat five classical baselines (Lasso, PCA, Random Forest importance, Mutual Information, and RFE) on accuracy while cutting the feature count by 25-60%. Five-fold cross-validation showed the gains were consistent rather than the result of a lucky split. We see this work as a small but concrete step toward integrating quantum methods into the full machine learning workflow, not just the classification stage.
Hybrid quantum-classical algorithms represent the dominant computational paradigm for near-term quantum advantage: classical co-processors handle data pre-processing, parameter optimisation, and post-processing, while quantum processors execute short-depth circuits that exploit superposition and entanglement for tasks classically hard to simulate. Despite growing interest, a systematic framework for designing, benchmarking, and selecting hybrid architectures across diverse computational task categories has been absent. This paper proposes the Hybrid Quantum-Classical Algorithm Design (HQCAD) framework, covering four hybrid paradigms: Variational Quantum-Classical Loops (VQCL) for optimisation and ground-state estimation; Quantum-Enhanced Machine Learning (QEML) combining quantum feature maps with classical classifiers; Quantum-Classical Divide-and-Conquer (QCDC) for graph and combinatorial problems; and Quantum-Accelerated Monte Carlo (QAMC) for sampling-based inference. HQCAD is benchmarked on six tasks -- molecular ground-state energy, binary classification, graph partitioning, financial Monte Carlo, protein fold scoring, and logistics routing -- on IBM Quantum Eagle (127-qubit) and IonQ Aria (25-qubit). VQCL achieves chemical accuracy (error < 1.6 mHa) for H2O ground-state energy. QEML achieves test AUC = 0.924 vs. classical SVM 0.884 on a quantum-advantage dataset. QAMC delivers 8.4x variance reduction for financial option pricing vs. classical Monte Carlo at equal shot budget. HQCAD provides a task-adaptive hybrid algorithm selection methodology validated across six computational domains.