Experimental Boson Sampling

TL;DR

This paper demonstrates a boson sampling experiment using integrated quantum networks, showcasing a promising intermediate quantum computing approach that can outperform classical computers with relatively few photons.

Contribution

It provides the first experimental realization of boson sampling with integrated photonic devices, validating the theoretical model and setting a benchmark for quantum computational advantage.

Findings

Successful implementation of three-photon interference in integrated devices

Experimental characterization confirms the boson-sampling output distribution

Sets a new benchmark for quantum computational experiments

Abstract

Universal quantum computers promise a dramatic speed-up over classical computers but a full-size realization remains challenging. However, intermediate quantum computational models have been proposed that are not universal, but can solve problems that are strongly believed to be classically hard. Aaronson and Arkhipov have shown that interference of single photons in random optical networks can solve the hard problem of sampling the bosonic output distribution which is directly connected to computing matrix permanents. Remarkably, this computation does not require measurement-based interactions or adaptive feed-forward techniques. Here we demonstrate this model of computation using high--quality laser--written integrated quantum networks that were designed to implement random unitary matrix transformations. We experimentally characterize the integrated devices using an in--situ reconstruction method and observe three-photon interference that leads to the boson-sampling output distribution. Our results set a benchmark for quantum computers, that hold the potential of outperforming conventional ones using only a few dozen photons and linear-optical elements.