Benchmarking integrated linear-optical architectures for quantum information processing

Flamini, Fulvio; Spagnolo, Nicolò; Viggianiello, Niko; Crespi, Andrea; Osellame, Roberto; Sciarrino, Fabio

doi:10.1038/s41598-017-15174-2

Cited by 46 publications

(68 citation statements)

References 56 publications

(75 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…is the realistic reflectivity, and t, r are the ideal transmissivity and reflectivity defined in Equation 2. The unitary matrices in Equation 5 cannot express the full transmissivity range of the MZI, with errors of up to 2 in the transmissivity, potentially limiting the performance of greedy progressive photonic algorithms [24][25][26]. Our Haar phase theory, which we develop in the following section, determines acceptable interferometer tolerances for calibration of a "perfect mesh" consisting of imperfect beamsplitters [21] given large N .…”

Section: B Beamsplitter Error Tolerancesmentioning

confidence: 99%

Matrix Optimization on Universal Unitary Photonic Devices

et al. 2019

View full text Add to dashboard Cite

Universal unitary photonic devices can apply arbitrary unitary transformations to a vector of input modes and provide a promising hardware platform for fast and energy-efficient machine learning using light. We simulate the gradient-based optimization of random unitary matrices on universal photonic devices composed of imperfect tunable interferometers. If device components are initialized uniform-randomly, the locally-interacting nature of the mesh components biases the optimization search space towards banded unitary matrices, limiting convergence to random unitary matrices. We detail a procedure for initializing the device by sampling from the distribution of random unitary matrices and show that this greatly improves convergence speed. We also explore mesh architecture improvements such as adding extra tunable beamsplitters or permuting waveguide layers to further improve the training speed and scalability of these devices.

show abstract

Section: B Beamsplitter Error Tolerancesmentioning

confidence: 99%

Matrix Optimization on Universal Unitary Photonic Devices

et al. 2019

View full text Add to dashboard Cite

show abstract

“…In addition, there will inevitably be fabrication imperfections of the photonic components (star couplers, waveguides, phase shifters, beamsplitters etc.) which break the correspondence between the trained software model and the hardware implementation [28], [56]. Such additional uncertainty introduced by physical implementation becomes non-negligible especially when scaling to a large number of components, hence it is important to evaluate its effects on the photonic CNN performance.…”

Section: B Non-idealities and Fabrication Imperfectionsmentioning

confidence: 99%

“…To study the effects of such imperfections, we evaluated the degradation of the MNIST classification accuracy of a pre-trained PCNN-784 (phase, linear) by introducing both amplitude and phase gaussian noise to the star coupler matrices F k and the complex-valued filter masks A k . We avoided adding noise to the fully-connected layers as this has been studied in previous literature along with several methods to ameliorate its effects being suggested [28], [56]. The added phase noise ∆φ is zero mean normally distributed with width 2πσ, whereas the amplitude noise is modeled as an additional loss i.e.…”

Section: B Non-idealities and Fabrication Imperfectionsmentioning

confidence: 99%

“…Previous studies have considered phase noise of width up to 0.02 rad, which is justified for low index contrast platforms [28], [56]. However, for high index contrast platforms like silicon-on-insulator, the phase errors resulting from imperfections could be up to 2 orders of magnitude greater [57] and hence we considered much larger phase errors.…”

Section: B Non-idealities and Fabrication Imperfectionsmentioning

confidence: 99%

See 1 more Smart Citation

Photonic convolutional neural networks using integrated diffractive optics

Ong¹,

Ang²,

Ooi³

et al. 2020

Preprint

View full text Add to dashboard Cite

With recent rapid advances in photonic integrated circuits, it has been demonstrated that programmable photonic chips can be used to implement artificial neural networks. Convolutional neural networks (CNN) are a class of deep learning methods that have been highly successful in applications such as image classification and speech processing. We present an architecture to implement a photonic CNN using the Fourier transform property of integrated star couplers. We show, in computer simulation, high accuracy image classification using the MNIST dataset. We also model component imperfections in photonic CNN and show that the performance degradation can be recovered in a programmable chip. Our proposed architecture provides a large reduction in physical footprint compared to current implementations as it utilizes the natural advantages of optics and hence offers a scalable pathway towards integrated photonic deep learning processors.

show abstract

“…However, Ref. [35] shows that integrated photonic architectures affected by standard fabrication and and propagation losses would allow for as many steps of a walk-like dynamics as 32 with only insignificant effects on the quality of the output walker state. The degree of reconfigurability of the large integrated-waveguides arrays is currently very limited, which would put some constraints to the possibility of simulating the effects of noise.…”

Section: Experimental Proposalmentioning

confidence: 99%

Robust quantum state engineering through coherent localization in biased-coin quantum walks

Majury

Boutari

O’Sullivan

et al. 2018

EPJ Quantum Technol.

View full text Add to dashboard Cite

We address the performance of a coin-biased quantum walk as a generator for non-classical position states of the walker. We exploit a phenomenon of coherent localization in the position space -resulting from the choice of small values of the coin parameter and assisted by post-selection -to engineer large-size coherent superpositions of position states of the walker. The protocol that we design appears to be remarkably robust against both the actual value taken by the coin parameter and strong dephasing-like noise acting on the spatial degree of freedom. We finally illustrate a possible linear-optics implementation of our proposal, suitable for both bulk and integrated-optics platforms.Quantum walks, which generalize the well-known concept of random walks to the quantum domain, have recently attracted considerable attention in light of the possibility that they offer the capability to explore non-trivial phenomena in statistical mechanics [1], such as interference [2], localization [3-6] and tunnelling, and are able to establish entanglement among either the different degrees of freedom of a walker, or the parties of a multi-walker setting [7][8][9][10][11][12][13]. Quantum walk-based architectures for quantum computation have been proposed and theoretically explored [14], while the possibility to implement quantum simulation through schemes based on the dynamics of a quantum walker are currently being pursued both theoretically and experimentally.In the last few years, the number of experimental validations of the quantum walk paradigm, and investigations towards its use for the coherent manipulation of information at the quantum mechanical level have flourished [15][16][17][18][19][20][21][22][23][24][25][26][27][28][29]. This therefore paves the way to the full exploitation of the possibilities offered by quantum walks for quantum state engineering in Hilbert spaces of large dimensions, which is in general a difficult task to pursue experimentally.In this paper, we explore exactly such a possibility and exploit the statistical features of a discrete-time quantum walk to engineer non-classical states of an N -dimensional system. Specifically, we make use of special coherent localization effects induced on the position state of a quantum walker by choosing properly the operation describing the tossing of a coin. We demonstrate that high-quality coherent superpositions of fully position states © The Author(s) 2018. This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.

show abstract

Benchmarking integrated linear-optical architectures for quantum information processing

Cited by 46 publications

References 56 publications

Matrix Optimization on Universal Unitary Photonic Devices

Matrix Optimization on Universal Unitary Photonic Devices

Photonic convolutional neural networks using integrated diffractive optics

Robust quantum state engineering through coherent localization in biased-coin quantum walks

Contact Info

Product

Resources

About