Abstract
Variational quantum algorithms (VQAs) provide a promising approach to achieving quantum advantage for practical problems on near-term noisy intermediate-scale quantum (NISQ) devices. Thus far, intensive studies on qubit-based VQAs have been made theoretically and experimentally on several physical platforms. However, there have been much fewer theoretical proposals and no experimental implementations on continuous-variable (CV) VQAs, although CV quantum computing can process infinite-dimensional quantum information even on single-mode devices and thus has great potential in the NISQ era. Here, we implement the CV version of one of the most typical VQAs, a quantum approximate optimization algorithm, on a single-mode programmable photonic quantum computer. We experimentally demonstrate that this algorithm solves a minimization problem of a given continuous real-valued function by implementing the quantum version of gradient descent and localizing an initially broadly-distributed wavefunction to the minimum of the given function. To the best of our knowledge, this is the first demonstration ever of a practical CV quantum algorithm on any physical platform, except for Gaussian Boson sampling. Our work highlights the power of CV quantum computing in the NISQ era, opening a new door to the quantum advantage in practical problems.
Abstract (translated)
URL
https://arxiv.org/abs/2206.07214