Multi-task learning (MTL) improves generalization and data efficiency by jointly learning related tasks through shared representations. In the widely used hard-parameter-sharing setting, a shared backbone is combined with task-specific prediction heads. However, task-specific parameters can grow rapidly with the number of tasks. Therefore, designing multi-task heads that preserve task specialization while improving parameter efficiency remains a key challenge. In Quantum Machine Learning (QML), variational quantum circuits (VQCs) provide a compact mechanism for mapping classical data to quantum states residing in high-dimensional Hilbert spaces, enabling expressive representations within constrained parameter budgets. We propose a parameter-efficient quantum multi-task learning (QMTL) framework that replaces conventional task-specific linear heads with a fully quantum prediction head in a hybrid architecture. The model consists of a VQC with a shared, task-independent quantum encoding stage, followed by lightweight task-specific ansatz blocks enabling localized task adaptation while maintaining compact parameterization. Under a controlled and capacity-matched formulation where the shared representation dimension grows with the number of tasks, our parameter-scaling analysis demonstrates that a standard classical head exhibits quadratic growth, whereas the proposed quantum head parameter cost scales linearly. We evaluate QMTL on three multi-task benchmarks spanning natural language processing, medical imaging, and multimodal sarcasm detection, where we achieve performance comparable to, and in some cases exceeding, classical hard-parameter-sharing baselines while consistently outperforming existing hybrid quantum MTL models with substantially fewer head parameters. We further demonstrate QMTL's executability on noisy simulators and real quantum hardware, illustrating its feasibility.
Multi-task learning (MTL) improves generalization and data efficiency by jointly learning related tasks through shared representations. In the widely used hard-parameter-sharing setting, a shared backbone is combined with task-specific prediction heads. However, task-specific parameters can grow rapidly with the number of tasks. Therefore, designing multi-task heads that preserve task specialization while improving parameter efficiency remains a key challenge. In Quantum Machine Learning (QML), variational quantum circuits (VQCs) provide a compact mechanism for mapping classical data to quantum states residing in high-dimensional Hilbert spaces, enabling expressive representations within constrained parameter budgets. We propose a parameter-efficient quantum multi-task learning (QMTL) framework that replaces conventional task-specific linear heads with a fully quantum prediction head in a hybrid architecture. The model consists of a VQC with a shared, task-independent quantum encoding stage, followed by lightweight task-specific ansatz blocks enabling localized task adaptation while maintaining compact parameterization. Under a controlled and capacity-matched formulation where the shared representation dimension grows with the number of tasks, our parameter-scaling analysis demonstrates that a standard classical head exhibits quadratic growth, whereas the proposed quantum head parameter cost scales linearly. We evaluate QMTL on three multi-task benchmarks spanning natural language processing, medical imaging, and multimodal sarcasm detection, where we achieve performance comparable to, and in some cases exceeding, classical hard-parameter-sharing baselines while consistently outperforming existing hybrid quantum MTL models with substantially fewer head parameters. We further demonstrate QMTL's executability on noisy simulators and real quantum hardware, illustrating its feasibility.
Quantum-centric supercomputing presents a compelling framework for large-scale hybrid quantum-classical tasks. Although quantum machine learning (QML) offers theoretical benefits in various applications, challenges such as large-size data encoding in the input stage and the reliance on quantum resources in the inference stage limit its practicality for tasks like fine-tuning large language models (LLMs). Quantum parameter generation, a novel approach of QML, addresses these limitations by using quantum neural networks (QNNs) to generate classical model weights (parameters) exclusively during training, thereby decoupling inference from quantum hardware. In this work, we introduce Quantum Parameter Adaptation (QPA) in the framework of quantum parameter generation, which integrates QNNs with a classical multi-layer perceptron mapping model to generate parameters for fine-tuning methods. Using Gemma-2 and GPT-2 as case studies, QPA demonstrates significant parameter reduction for parameter-efficient fine-tuning methods, such as Low-Rank Adaptation (LoRA), while maintaining comparable or improved performance in text generation tasks. Specifically, QPA reduces the number of parameters to 52.06% of the original LoRA for GPT-2 with a slight performance gain of 0.75%, and to 16.84% for Gemma-2, with a marginal performance improvement of 0.07%. These results highlight QPA's ability to achieve efficient parameter reduction without sacrificing performance in the quantum parameter generation framework. This work showcases the potential of quantum-enhanced parameter reduction, offering a scalable quantum-classical solution for fine-tuning LLMs while preserving the feasibility of inference on classical hardware.
Medical image classification is a critical task in computer-aided diagnostics, but classical deep learning models often demand substantial computational resources and large parameter spaces. This paper investigates the potential of hybrid quantum-classical neural networks to address this challenge in the context of multi-class brain tumor classification from MRI scans. We introduce a custom hybrid quantum machine learning (QML) model and conduct a rigorous comparative analysis against its purely classical counterpart, as well as established architectures such as ResNet-18 and VGG-16. Our findings demonstrate that the proposed hybrid QML model achieves a state-of-the-art test accuracy of 96.58%, matching the performance of the significantly larger ResNet-18 model. Crucially, the hybrid model outperforms its direct classical counterpart, the Classical Benchmark (96.20%), while utilizing approximately 80% fewer parameters. Although simulating quantum circuits results in longer training times, our findings highlight the remarkable parameter efficiency and competitive accuracy of hybrid quantum models, suggesting a promising direction for developing powerful yet lightweight networks for medical diagnostics.
Abstract Variational quantum circuits (VQCs) are a central paradigm for quantum machine learning, yet their practical deployment is limited by poor trainability, sensitivity to noise, and the lack of input-adaptive parameterization. In this work, we introduce TensorDual-VQC, a tensor-structured operator-learning framework that models variational quantum computation as an input-conditioned mapping that jointly governs quantum state encoding and circuit parameter generation. The proposed approach employs a shared low-rank Tensor-Train (TT) backbone with dual output heads, inducing a coupled representation that constrains the hypothesis space to a structured low-dimensional manifold. Under standard smoothness and low-rank assumptions, we show that the approximation behavior is governed by the TT-rank rather than the ambient parameter dimension, and that shared latent structure can enable more compact representations than decoupled parameterizations in structured regimes. We further analyze trainability through the induced Neural Tangent Kernel, showing that the multi-branch gradient structure enriches the span of gradient features and can improve kernel conditioning under mild non-degeneracy conditions. In addition, we relate robustness to quantum noise to the intrinsic effective dimension of the tensor parameter manifold, providing a first-order scaling characterization of output variance under stochastic perturbations. We evaluate TensorDual-VQC on quantum dot classification, quantum simulation, and combinatorial optimization tasks under both simulated noise and real quantum hardware. Across all settings, the proposed framework achieves higher accuracy, faster convergence, and greater stability than standard VQCs and hypernetwork-based baselines when quantum resources are matched. These results suggest that structured operator learning with shared low-rank representations provides a principled and practical approach to scalable, noise-resilient quantum learning in the NISQ regime.
Recent advances in quantum computing and machine learning have given rise to quantum machine learning (QML), with growing interest in learning from sequential data. Quantum recurrent models like QLSTM are promising for time-series prediction, NLP, and reinforcement learning. However, designing effective variational quantum circuits (VQCs) remains challenging and often task-specific. To address this, we propose DiffQAS-QLSTM, an end-to-end differentiable framework that optimizes both VQC parameters and architecture selection during training. Our results show that DiffQAS-QLSTM consistently outperforms handcrafted baselines, achieving lower loss across diverse test settings. This approach opens the door to scalable and adaptive quantum sequence learning.
Hybrid quantum-classical learning models increasingly integrate neural networks with variational quantum circuits (VQCs) to exploit complementary inductive biases. However, many existing approaches rely on tightly coupled architectures or task-specific encoders, limiting conceptual clarity, generality, and transferability across learning settings. In this work, we introduce Quantum LEGO Learning, a modular and architecture-agnostic learning framework that treats classical and quantum components as reusable, composable learning blocks with well-defined roles. Within this framework, a pre-trained classical neural network serves as a frozen feature block, while a VQC acts as a trainable adaptive module that operates on structured representations rather than raw inputs. This separation enables efficient learning under constrained quantum resources and provides a principled abstraction for analyzing hybrid models. We develop a block-wise generalization theory that decomposes learning error into approximation and estimation components, explicitly characterizing how the complexity and training status of each block influence overall performance. Our analysis generalizes prior tensor-network-specific results and identifies conditions under which quantum modules provide representational advantages over comparably sized classical heads. Empirically, we validate the framework through systematic block-swap experiments across frozen feature extractors and both quantum and classical adaptive heads. Experiments on quantum dot classification demonstrate stable optimization, reduced sensitivity to qubit count, and robustness to realistic noise.
Accurate and computationally efficient classification of brain tumors from MRI scans is critical for deployment in resource-constrained clinical settings, where conventional deep CNN architectures-despite achieving strong accuracy-often require large parameter counts that limit scalability. This work proposes a hybrid quantum-classical deep learning framework that combines a classical convolutional feature extractor with a variational quantum circuit (VQC) for multi-class tumor classification. Spatial features extracted by the CNN backbone will be dimensionality-reduced and encoded into quantum states using angle encoding; a parameterized quantum circuit will then process these features in a higher-dimensional Hilbert space, leveraging superposition and entanglement to capture non-linear feature correlations. Measurement outputs from the quantum layer will be passed to a classical fully connected layer for classification into glioma, meningioma, pituitary tumor, and no-tumor categories. The proposed framework will be evaluated on a publicly available benchmark MRI dataset and benchmarked against state-of-the-art classical CNN architectures in terms of accuracy, precision, recall, F1-score, and parameter efficiency. This study aims to demonstrate that quantum-enhanced layers can achieve competitive classification performance while substantially reducing trainable parameters, offering a practical direction for efficient, quantum-assisted medical image analysis.