Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Digital optical phase and amplitude matrix multiplication processor for neural networks

Xiansong Meng, Kwangwoong Kim, Po Dong, Deming Kong, Hao Hu

Published Jan 1, 2024• 2 citations
Publisher

Abstract

We propose a high-precision digital optical matrix multiplier utilizing phase and amplitude for neural networks. Results show error-free performance with 16-bit precision in high-definition image processing and no accuracy loss in handwritten digit recognition task.
DOI
10.1364/cleo_si.2024.sth5c.8

Related papers

Multidimensionally-Encoded High-Precision Optical Multiplier for Matrix Multiplication

Pierre Nay, Xiansong Meng, Mingyang Gao +4 more

We propose and demonstrate a proof-of-concept optical multiplier for matrix multiplication using multidimensional encoding to increase the numerical precision. We achieve error-free performance for 8-bit image processing, and a low error rate of$8.76 \times 10^{-5}$for 8-bit random number multiplication.

European Conference on Optical Communication•Sep 28, 2025•0 citations

Error codes applied to optical algebraic processors

Scott A. Ellett, John F. Walkup, Thomas F. Krile

Optical algebraic processors can perform complex calculations in parallel and at high speeds. However, they commonly suffer from a low analog accuracy which hinders their widespread application. Error detection and correction codes provide one technique for improving the accuracy of optical algebraic processors. The use of these codes would allow some of the errors that may occur during a computation to be detected and possibly corrected. This paper describes the results of various computer simulations of optical matrix-vector multipliers employing error-correction codes. It discusses the application of convolutional codes to optical matrix-vector multipliers along with several block codes. Both binary and nonbinary codes are considered. The results indicate that a significant improvement in performance can be obtained when compared with processors not employing error-correction codes. Also, the type of noise, whether signal-independent or signal-dependent noise, has a significant effect on the performance of a matrix-vector multiplier employing an error code. The encoding and decoding operations required for the error codes can be performed optically.

Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIE•Nov 1, 1991•1 citation

Photonic-computing error correction through optical en-/decoder calibrations

Adam Carstensen, Babak Vosoughi Lahijani

Photonic processors have emerged as an attractive platform for fast and energy-efficient matrix-vector multiplication. However, they are susceptible to error due to their analog nature. Here, we present an error-correction technique that implements a correction offset to the optical en-/decoders of photonic processors. Our proposed method is general-purpose, does not require introducing any additional components to the photonic network, and can address errors stemming from unbalanced losses, 50/50 beamsplitter deviations, digital-to-analog conversion inaccuracies, and any unknown sources. In particular, we show that our method is highly effective in mitigating unbalanced-loss errors, a problem that has not previously been addressed by any error-correction technique. Using this approach, we achieve over 90% error reduction in large triangular meshes, overcoming a key obstacle to highly accurate photonic processors for information processing.

arXiv:2602.18299•Feb 20, 2026•0 citations

An optical neural network operating with less than 1 photon per multiplication

Tianyu Wang, Shi-Yuan Ma, Logan G. Wright +3 more

We report an experimental demonstration of an optical neural network performing image classification with high accuracy using less than a single photon per scalar multiplication, validating theoretical predictions about quantum-limited performance of ONNs.

Frontiers in Optics•Jan 1, 2021•4 citations

Hybrid quantum-classical photonic neural networks

Tristan Austin, Simon Bilodeau, Andrew Hayman +2 more

Neuromorphic (brain-inspired) photonics accelerates AI 1 with high-speed, energy-efficient solutions for RF communication 2 , image processing 3 , 4 , and fast matrix multiplication 5 , 6 . However, integrated neuromorphic photonic hardware faces size constraints that limit network complexity. Recent advances in photonic quantum hardware 7 and performant trainable quantum circuits 8 offer a path to more scalable photonic neural networks. Here, we show that a combination of classical network layers with trainable continuous variable quantum circuits yields hybrid networks with improved trainability and accuracy. On a classification task, these hybrid networks match the performance of classical networks nearly twice their size. These performance benefits remain even when evaluated at state-of-the-art bit precisions for classical and quantum hardware. Finally, we outline available hardware and a roadmap to hybrid architectures. These hybrid quantum-classical networks demonstrate a unique route to enhance the computational capacity of integrated photonic neural networks without increasing the network size.

npj Unconventional Computing•Dec 1, 2025•3 citations

A fully-programmable integrated photonic processor for both domain-specific and general-purpose computing

Feng-Kai Han, Xiaoyun Xu, Tianyu Zhang +10 more

A variety of complicated computational scenarios have made unprecedented demands on the computing power and energy efficiency of electronic computing systems, including solving intractable nondeterministic polynomial-time (NP)-complete problems and dealing with large-scale artificial intelligence models. Optical computing emerges as a promising paradigm to meet these challenges, whereas current optical computing architectures have limited versatility. Their applications are usually either constrained to a specialized domain or restricted to general-purpose matrix computation. Here, we implement a fully-programmable integrated photonic processor that can be configured to tackle both specific computational problems and general-purpose matrix computation. We achieve complete end-to-end control of the photonic processor by utilizing a self-developed integrated programmable optoelectronic computing platform. For domain-specific computing, our photonic processor can efficiently solve two kinds of NP-complete problems: subset sum problem (far more than 2^N different instances) and exact cover problem. For general-purpose computation, we experimentally demonstrate high-precision optical dot product and further realize accurate image edge detection and MNIST handwritten image classification task with an accuracy of 97%. Our work enhances the versatility and capability of optical computing architecture, paving the way for its practical application in future high-performance and complex computing scenarios.

arXiv:2508.13551•Aug 19, 2025•0 citations
We propose and demonstrate a proof-of-concept optical multiplier for matrix multiplication using multidimensional encoding to increase the numerical precision. We achieve error-free performance for 8-bit image processing, and a low error rate of8.76×10−58.76 \times 10^{-5}8.76×10−5for 8-bit random number multiplication.
Optical algebraic processors can perform complex calculations in parallel and at high speeds. However, they commonly suffer from a low analog accuracy which hinders their widespread application. Error detection and correction codes provide one technique for improving the accuracy of optical algebraic processors. The use of these codes would allow some of the errors that may occur during a computation to be detected and possibly corrected. This paper describes the results of various computer simulations of optical matrix-vector multipliers employing error-correction codes. It discusses the application of convolutional codes to optical matrix-vector multipliers along with several block codes. Both binary and nonbinary codes are considered. The results indicate that a significant improvement in performance can be obtained when compared with processors not employing error-correction codes. Also, the type of noise, whether signal-independent or signal-dependent noise, has a significant effect on the performance of a matrix-vector multiplier employing an error code. The encoding and decoding operations required for the error codes can be performed optically.
Photonic processors have emerged as an attractive platform for fast and energy-efficient matrix-vector multiplication. However, they are susceptible to error due to their analog nature. Here, we present an error-correction technique that implements a correction offset to the optical en-/decoders of photonic processors. Our proposed method is general-purpose, does not require introducing any additional components to the photonic network, and can address errors stemming from unbalanced losses, 50/50 beamsplitter deviations, digital-to-analog conversion inaccuracies, and any unknown sources. In particular, we show that our method is highly effective in mitigating unbalanced-loss errors, a problem that has not previously been addressed by any error-correction technique. Using this approach, we achieve over 90% error reduction in large triangular meshes, overcoming a key obstacle to highly accurate photonic processors for information processing.
Neuromorphic (brain-inspired) photonics accelerates AI 1 with high-speed, energy-efficient solutions for RF communication 2 , image processing 3 , 4 , and fast matrix multiplication 5 , 6 . However, integrated neuromorphic photonic hardware faces size constraints that limit network complexity. Recent advances in photonic quantum hardware 7 and performant trainable quantum circuits 8 offer a path to more scalable photonic neural networks. Here, we show that a combination of classical network layers with trainable continuous variable quantum circuits yields hybrid networks with improved trainability and accuracy. On a classification task, these hybrid networks match the performance of classical networks nearly twice their size. These performance benefits remain even when evaluated at state-of-the-art bit precisions for classical and quantum hardware. Finally, we outline available hardware and a roadmap to hybrid architectures. These hybrid quantum-classical networks demonstrate a unique route to enhance the computational capacity of integrated photonic neural networks without increasing the network size.
A variety of complicated computational scenarios have made unprecedented demands on the computing power and energy efficiency of electronic computing systems, including solving intractable nondeterministic polynomial-time (NP)-complete problems and dealing with large-scale artificial intelligence models. Optical computing emerges as a promising paradigm to meet these challenges, whereas current optical computing architectures have limited versatility. Their applications are usually either constrained to a specialized domain or restricted to general-purpose matrix computation. Here, we implement a fully-programmable integrated photonic processor that can be configured to tackle both specific computational problems and general-purpose matrix computation. We achieve complete end-to-end control of the photonic processor by utilizing a self-developed integrated programmable optoelectronic computing platform. For domain-specific computing, our photonic processor can efficiently solve two kinds of NP-complete problems: subset sum problem (far more than 2^N different instances) and exact cover problem. For general-purpose computation, we experimentally demonstrate high-precision optical dot product and further realize accurate image edge detection and MNIST handwritten image classification task with an accuracy of 97%. Our work enhances the versatility and capability of optical computing architecture, paving the way for its practical application in future high-performance and complex computing scenarios.