We introduce a hybrid quantum-classical adversarial machine learning architecture called a quantum-classical associative adversarial network (QAAN). This architecture consists of a classical generative adversarial network with a small auxiliary quantum Boltzmann machine that is simultaneously trained on an intermediate layer of the discriminator of the generative network. We numerically study the performance of QAANs compared to their classical counterparts on the MNIST and CIFAR-10 data sets, and show that QAANs attain a higher quality of learning when evaluated using the Inception score and the Fr\'echet Inception distance. As the QAAN architecture only relies on sampling simple local observables of a small quantum Boltzmann machine, this model is particularly amenable for implementation on the current and next generations of quantum devices.
We introduce a hybrid quantum-classical adversarial machine learning architecture called a quantum-classical associative adversarial network (QAAN). This architecture consists of a classical generative adversarial network with a small auxiliary quantum Boltzmann machine that is simultaneously trained on an intermediate layer of the discriminator of the generative network. We numerically study the performance of QAANs compared to their classical counterparts on the MNIST and CIFAR-10 data sets, and show that QAANs attain a higher quality of learning when evaluated using the Inception score and the Fr'echet Inception distance. As the QAAN architecture only relies on sampling simple local observables of a small quantum Boltzmann machine, this model is particularly amenable for implementation on the current and next generations of quantum devices.
Abstract Generative models have the capacity to model and generate new examples from a dataset and have an increasingly diverse set of applications driven by commercial and academic interest. In this work, we present an algorithm for learning a latent variable generative model via generative adversarial learning where the canonical uniform noise input is replaced by samples from a graphical model. This graphical model is learned by a Boltzmann machine which learns low-dimensional feature representation of data extracted by the discriminator. A quantum processor can be used to sample from the model to train the Boltzmann machine. This novel hybrid quantum-classical algorithm joins a growing family of algorithms that use a quantum processor sampling subroutine in deep learning, and provides a scalable framework to test the advantages of quantum-assisted learning. For the latent space model, fully connected, symmetric bipartite and Chimera graph topologies are compared on a reduced stochastically binarized MNIST dataset, for both classical and quantum sampling methods. The quantum-assisted associative adversarial network successfully learns a generative model of the MNIST dataset for all topologies. Evaluated using the Fréchet inception distance and inception score, the quantum and classical versions of the algorithm are found to have equivalent performance for learning an implicit generative model of the MNIST dataset. Classical sampling is used to demonstrate the algorithm on the LSUN bedrooms dataset, indicating scalability to larger and color datasets. Though the quantum processor used here is a quantum annealer, the algorithm is general enough such that any quantum processor, such as gate model quantum computers, may be substituted as a sampler.
This paper introduces a novel hybrid quantumclassical model using QGAN (quantum generative adversarial network) to achieve more efficient, convergent and noiserobust machine learning for nearterm quantum (NISQ) applications. Unlike the existing GAN type models that were purely classical or purely quantum, to propose a quantum generator consisting of shallow variation circuits (48 qubits) together with a classical discriminator that is crucial for the stability of adversarial training. The empirical results on the MNIST dataset showed that were able to perform classification with an accuracy of 94.2%, an improvement of over 3.2% from a straightforward classical only approach, while NVGAN also improved its generative performance to a Fréchet Inception Distance (FID) score of 38.7 or 26% in quality enhancement from synthesis perspective. Both trainings converged 50 epochs earlier than each training when used the hybrid model (30% improvement in convergence). The key innovation of this work is bringing (1) to bear both the utility of the quantumenhanced feature representation to have at disposal, and (2) classical stability measures and erroraware optimizations (such as parameter clipping, early stopping), such that its hybridization enabled it to continue operating with>90%accuracy under cases of reasonable depolarizing/amplitude damping noise on an MLP where purely quantum formulations decayed to (approx 70%). Altogether, the paper validates the presented hybrid QGAN approach as a statistically substantiated noise robust and scalable representational tool for real world applications of quantum enhanced machine learning tasks on NISQ platforms.
Generative adversarial network (GAN) is one of the widely-adopted machine-learning frameworks for a wide range of applications such as generating high-quality images, video, and audio contents. However, training a GAN could become computationally expensive for large neural networks. In this work, we propose a hybrid quantum-classical architecture for improving GAN (denoted as QC-GAN). The performance was examed numerically by benchmarking with a classical GAN using MindSpore Quantum on the task of hand-written image generation. The generator of the QC-GAN consists of a quantum variational circuit together with a one-layer neural network, and the discriminator consists of a traditional neural network. Leveraging the entangling and expressive power of quantum circuits, our hybrid architecture achieved better performance (Frechet Inception Distance) than the classical GAN, with much fewer training parameters and number of iterations for convergence. We have also demonstrated the superiority of QC-GAN over an alternative quantum GAN, namely pathGAN, which could hardly generate 16$\times$16 or larger images. This work demonstrates the value of combining ideas from quantum computing with machine learning for both areas of Quantum-for-AI and AI-for-Quantum.
Abstract Quantum generative adversarial networks (QGANs) have demonstrated strong capabilities in tasks like synthetic data generation and detecting anomalies. Recent developments have increasingly integrated traditional machine learning techniques to boost the performance of QGANs. Motivated by this progress, we propose an innovative QGAN architecture that incorporates a classical learning component and employs a dual-generator design. Our approach improves upon the traditional hybrid quantum–classical GAN structure and introduces a redesigned loss function tailored for the new model. Experiments on multiple datasets indicate that our method surpasses previous techniques in image generation quality, achieving a 1.38% average reduction in Fréchet inception distance scores compared to the current state-of-the-art, and improvements of 6.52%, 0.36%, and 0.38% in structural similarity index, cosine similarity, and peak signal-to-noise ratio metrics, respectively. Additionally, our architecture supports the generation of larger images (up to 78 × 78 ), as verified on the CelebA dataset. Simulations conducted in noisy conditions further confirm the robustness and effectiveness of both the proposed architecture and loss function.
This paper presents a novel framework for Quantum Generative Adversarial Networks (QGANs) that integrates Quantum Kernel Learning into the discriminator architecture. By replacing the classical discriminator with a Quantum Kernel Discriminator (QKD), the proposed model leverages high-dimensional Hilbert space mappings to enhance separability between real and generated data. The quantum discriminator employs Pauli Feature Maps and Fidelity Quantum Kernels to evaluate sample similarity in the quantum space, leading to improved training stability and reduced mode collapse. Empirical results on the MNIST dataset demonstrate faster convergence, higher fooling scores, and improved sample diversity compared to classical GANs. Additionally, the application of Quantum Kernel Methods to the Iris dataset highlights the generalization capabilities of the QKD approach. These findings establish the potential of hybrid quantum-classical architectures in advancing generative modeling and open new avenues for applying quantum machine learning in low-data and high-complexity domains.
Quantum Generative Adversarial Networks (QGANs), as a rising paradigm in Quantum Machine Learning, have shown promising potential in image generation and processing. However, their output quality remains suboptimal, and existing research is largely limited to small-scale, proof-of-concept studies. In this work, we propose a hybrid quantum-classical GAN architecture, where the generator integrates parameterized quantum circuits (PQCs) and classical neural networks. This integration significantly enhances the visual quality of generated images. Our model leverages the complementary strengths of quantum and classical components and outperforms existing methods (Tsang et al. in IEEE Trans. Quantum Eng. 4:1–19, 2023 ; Gulrajani et al. in Proceedings of the 31st international conference on neural information processing systems. NIPS’17, Red Hook, pp. 5769–5779, 2017 ), particularly in terms of image fidelity. Experiments conducted on the MNIST family of datasets show that our hybrid approach achieves a 20.26% average reduction in Fréchet Inception Distance. Furthermore, it improves the Structural Similarity Index Measure, Cosine Similarity, and Peak Signal-to-Noise Ratio by 26.04%, 2.22%, and 7.62%, respectively. These results highlight the effectiveness of combining quantum computing with machine learning, and underscore the potential of hybrid quantum-classical models in advancing generative tasks.