Learning and inference on generative adversarial quantum circuits

Abstract

Quantum mechanics is inherently probabilistic in light of Born's rule. Using quantum circuits as probabilistic generative models for classical data exploits their superior expressibility and efficient direct sampling ability. However, training of quantum circuits can be more challenging compared to classical neural networks due to lack of efficient differentiable learning algorithm. We devise an adversarial quantum-classical hybrid training scheme via coupling a quantum circuit generator and a classical neural network discriminator together. After training, the quantum circuit generative model can infer missing data with quadratic speed up via amplitude amplification. We numerically simulate the learning and inference of generative adversarial quantum circuit using the prototypical Bars-and-Stripes dataset. Generative adversarial quantum circuits is a fresh approach to machine learning which may enjoy the practically useful quantum advantage on near-term quantum devices. arXiv:1808.03425v1 [quant-ph]

The Born supremacy: quantum advantage and training of an Ising Born machine

“…There is a slight difference between Eq. (17) and that of ref. 14 , due to a different parameterisation of the unitaries above.…”

“…The initial proposals to train QCBMs were gradientfree 15,31 , but gradient-based methods have also been proposed 14,16,32 . In this work, we advocate for increasing the classical computational power required in training to achieve better performance, rather than increasing the quantum resources, for example by adding extra ancillae 16 or adding costly and potentially unstable (quantum) adversaries 17,33,34 .…”

“…Specifically, for a state |ψ〉, a measurement produces a sample x~p(x) = |〈x|ψ〉| 2 . There are several variants, including Bayesian approaches 16 , adversarial training methods 17 , and adaptations to continuous distributions 18 .…”

Parameterized quantum circuits as machine learning models

“…We discuss the case of quantum data in the next Section while here we focus on classical data. Both Situ et al [93] and Zeng et al [94] couple a PQC generator to an ANN discriminator and successfully reproduce the statistics of discrete target distributions. Romero and Aspuru-Guzik [95] ex- In the quantum generative adversarial network the generator creates synthetic samples and the discriminator tries to distinguish between the generated and the real samples.…”

Generative model benchmarks for superconducting qubits

“…Generative models, such as adversarial networks [5], have recently spurred significant interest in the development of quantum circuit analogues [6,7] and adversarial quantum circuits training [8][9][10]. The quantum-circuit Born machine (QCBM) is a generative model constructed as a quantum circuit [3,4,11].…”