Source-linked AI summary

Quantum generative adversarial learning in a superconducting quantum circuit

Ling Hu, Shu-Hao Wu, Weizhou Cai, Yuwei Ma, Xianghao Mu, Yuan Xu, Haiyan Wang, Yipu Song, Dong-Ling Deng, Chang-Ling Zou, Luyan Sun

arXiv:1808.02893v1quant-phcond-mat.dis-nncond-mat.supr-con

TL;DR

This work addresses whether quantum generative adversarial learning can be demonstrated experimentally with quantum input and output data. It implements QGAN in a superconducting quantum circuit and shows that generated states replicate quantum-data statistics with high fidelity, while identifying measurement precision and experimental imperfections as practical constraints.

  • Problem

    Quantum generative adversarial learning had been theoretically proposed, but experimental demonstrations with quantum input and output data were still needed.

  • Method

    A superconducting-circuit QGAN uses a quantum generator to produce state ensembles and a quantum discriminator to distinguish them from arbitrary states generated by a digital qubit channel simulator.

  • Results

    98.8% average final-state fidelity was achieved for both pure and mixed quantum data across 100 random adversarial learning processes.

  • Takeaways & Limitations

    The experiment demonstrates feasible quantum generative adversarial learning on a superconducting quantum circuit, with a protocol suited to noisy intermediate-scale quantum devices.

  • Takeaways & Limitations

    Measurement precision, qubit decoherence, and non-ideal measured gradients constrain convergence and make experimental learning slower and less accurate than noiseless simulations.

Abstract

from arXiv · show

Generative adversarial learning is one of the most exciting recent breakthroughs in machine learning---a subfield of artificial intelligence that is currently driving a revolution in many aspects of modern society. It has shown splendid performance in a variety of challenging tasks such as image and video generations. More recently, a quantum version of generative adversarial learning has been theoretically proposed and shown to possess the potential of exhibiting an exponential advantage over its classical counterpart. Here, we report the first proof-of-principle experimental demonstration of quantum generative adversarial learning in a superconducting quantum circuit. We demonstrate that, after several rounds of adversarial learning, a quantum state generator can be trained to replicate the statistics of the quantum data output from a digital qubit channel simulator, with a high fidelity ($98.8\%$ on average) that the discriminator cannot distinguish between the true and the generated data. Our results pave the way for experimentally exploring the intriguing long-sought-after quantum advantages in machine learning tasks with noisy intermediate-scale quantum devices.

Loading 1808.02893v1…