Quantum generative adversarial networks

TL;DR

This paper introduces quantum generative adversarial networks, extending classical adversarial training to quantum circuits, demonstrating their trainability through a simple numerical experiment with potential applications in quantum machine learning.

Contribution

It presents the first framework for quantum generative adversarial networks and methods to compute gradients using quantum circuits.

Findings

Quantum GANs can be trained successfully in simple experiments.

Gradient computation can be performed using quantum circuits.

The approach paves the way for practical quantum machine learning applications.

Abstract

Quantum machine learning is expected to be one of the first potential general-purpose applications of near-term quantum devices. A major recent breakthrough in classical machine learning is the notion of generative adversarial training, where the gradients of a discriminator model are used to train a separate generative model. In this work and a companion paper, we extend adversarial training to the quantum domain and show how to construct generative adversarial networks using quantum circuits. Furthermore, we also show how to compute gradients -- a key element in generative adversarial network training -- using another quantum circuit. We give an example of a simple practical circuit ansatz to parametrize quantum machine learning models and perform a simple numerical experiment to demonstrate that quantum generative adversarial networks can be trained successfully.