Quantum Wasserstein Generative Adversarial Networks
This work addresses the challenge of improving quantum generative models for applications in quantum machine learning and chemistry, though it appears incremental as it adapts classical WGAN concepts to the quantum domain.
The authors tackled the problem of adversarial training for quantum generative models by proposing quantum Wasserstein Generative Adversarial Networks (WGANs), which demonstrated more robust and scalable performance in numerical simulations and generated a 3-qubit circuit with ~50 gates that approximates a circuit requiring over 10k gates.
The study of quantum generative models is well-motivated, not only because of its importance in quantum machine learning and quantum chemistry but also because of the perspective of its implementation on near-term quantum machines. Inspired by previous studies on the adversarial training of classical and quantum generative models, we propose the first design of quantum Wasserstein Generative Adversarial Networks (WGANs), which has been shown to improve the robustness and the scalability of the adversarial training of quantum generative models even on noisy quantum hardware. Specifically, we propose a definition of the Wasserstein semimetric between quantum data, which inherits a few key theoretical merits of its classical counterpart. We also demonstrate how to turn the quantum Wasserstein semimetric into a concrete design of quantum WGANs that can be efficiently implemented on quantum machines. Our numerical study, via classical simulation of quantum systems, shows the more robust and scalable numerical performance of our quantum WGANs over other quantum GAN proposals. As a surprising application, our quantum WGAN has been used to generate a 3-qubit quantum circuit of ~50 gates that well approximates a 3-qubit 1-d Hamiltonian simulation circuit that requires over 10k gates using standard techniques.