🤖 AI Summary
Quantum generative models remain in their infancy, lacking systematic methodologies and efficient mechanisms for leveraging quantum noise. Method: This paper proposes a hybrid quantum generative framework for image synthesis, wherein the generator is implemented via a variational quantum circuit. It innovatively integrates intrinsic hardware quantum noise with controllable temporal noise scheduling to enable synergistic modeling; further, it introduces an adaptive noise injection strategy embedded within a hybrid quantum-classical training architecture. Contribution/Results: Evaluated on MNIST and MedMNIST, the model substantially outperforms existing quantum generative baselines. Empirical results demonstrate that quantum noise can be deliberately engineered and regulated—not only enhancing generation fidelity but also establishing noise as a learnable resource, thereby pioneering a new paradigm in quantum generative modeling.
📝 Abstract
Research on quantum generative models is currently in its early exploratory stages, with very few established methodologies. In this paper, we propose a novel hybrid quantum generative model based on variational quantum circuits for image generation tasks, introducing innovative noise techniques specifically tailored for quantum computation. Our approach utilizes two distinctive noise strategies: quantum-generated noise inherent to quantum circuits, and a newly developed noise scheduling method, applying different noise levels strategically across time steps during the training process. Experiments conducted on MNIST and MedMNIST datasets demonstrate that our hybrid quantum model, combined with these specialized noise techniques, achieves promising results, suggesting improved generative performance compared to baseline quantum generative approaches. This exploratory work lays a critical foundation and opens new avenues for advancing quantum generative modeling research.