ReDON: Recurrent Diffractive Optical Neural Processor with Reconfigurable Self-Modulated Nonlinearity

• URL: http://arxiv.org/abs/2602.23616v2
• Date: Mon, 02 Mar 2026 19:28:21 GMT
• Title: ReDON: Recurrent Diffractive Optical Neural Processor with Reconfigurable Self-Modulated Nonlinearity
• Authors: Ziang Yin, Qi Jing, Raktim Sarma, Rena Huang, Yu Yao, Jiaqi Gu,
• Abstract summary: We introduce the Recurrent Diffractive Optical Neural Processor (ReDON), a novel architecture featuring reconfigurable, recurrent self-modulated nonlinearity.<n>Inspired by the gated linear unit (GLU) used in large language models, ReDON senses a fraction of the propagating optical field and modulates its phase or intensity via a lightweight parametric function.<n>On image recognition and segmentation benchmarks, ReDON improves test accuracy and mean intersection-over-union (mIoU) by up to 20% compared with prior DONNs.
• Score: 4.488347887618485
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Diffractive optical neural networks (DONNs) have demonstrated unparalleled energy efficiency and parallelism by processing information directly in the optical domain. However, their computational expressivity is constrained by static, passive diffractive phase masks that lack efficient nonlinear responses and reprogrammability. To address these limitations, we introduce the Recurrent Diffractive Optical Neural Processor (ReDON), a novel architecture featuring reconfigurable, recurrent self-modulated nonlinearity. This mechanism enables dynamic, input-dependent optical transmission through in-situ electro-optic self-modulation, providing a highly efficient and reprogrammable approach to optical computation. Inspired by the gated linear unit (GLU) used in large language models, ReDON senses a fraction of the propagating optical field and modulates its phase or intensity via a lightweight parametric function, enabling effective nonlinearity with minimal inference overhead. As a non-von Neumann architecture in which the primary weighting elements (metasurfaces) remain fixed, ReDON substantially extends the nonlinear representational capacity and task adaptability of conventional DONNs through recurrent optical hardware reuse and dynamically tunable nonlinearity. We systematically investigate various self-modulation configurations to characterize the trade-offs between hardware efficiency and computational expressivity. On image recognition and segmentation benchmarks, ReDON improves test accuracy and mean intersection-over-union (mIoU) by up to 20% compared with prior DONNs employing either optical or digital nonlinearities at comparable model complexity and negligible additional power consumption. This work establishes a new paradigm for reconfigurable nonlinear optical computing, uniting recurrence and self-modulation within non-von Neumann analog processors.

Related papers

• Gradient Descent as a Perceptron Algorithm: Understanding Dynamics and Implicit Acceleration [67.12978375116599]
  We show that the steps of gradient descent (GD) reduce to those of generalized perceptron algorithms.<n>This helps explain the optimization dynamics and the implicit acceleration phenomenon observed in neural networks.
  arXiv  Detail & Related papers  (2025-12-12T14:16:35Z)
• Spatially Parallel All-optical Neural Networks [14.284567977850912]
  All-optical neural networks (AONNs) have emerged as a promising paradigm for ultrafast and energy-efficient computation.<n>Here we propose a spatially parallel architecture for all-optical neural networks (SP-AONNs)<n>Our findings highlight spatial parallelism as a practical and scalable strategy for advancing the capabilities of optical neural computing.
  arXiv  Detail & Related papers  (2025-09-28T03:25:40Z)
• Massively parallel and universal approximation of nonlinear functions using diffractive processors [17.16859564691328]
  Large-scale nonlinear computation can be performed using linear optics through optimized diffractive processors composed of passive phase-only surfaces.<n>We numerically demonstrate the parallel computation of one million distinct nonlinear functions, accurately executed at wavelength-scale spatial density.<n>These results establish diffractive optical processors as a scalable platform for massively parallel universal nonlinear function approximation.
  arXiv  Detail & Related papers  (2025-07-11T01:54:10Z)
• Generative System Dynamics in Recurrent Neural Networks [56.958984970518564]
  We investigate the continuous time dynamics of Recurrent Neural Networks (RNNs)<n>We show that skew-symmetric weight matrices are fundamental to enable stable limit cycles in both linear and nonlinear configurations.<n> Numerical simulations showcase how nonlinear activation functions not only maintain limit cycles, but also enhance the numerical stability of the system integration process.
  arXiv  Detail & Related papers  (2025-04-16T10:39:43Z)
• Training Large-Scale Optical Neural Networks with Two-Pass Forward Propagation [0.0]
  This paper addresses the limitations in Optical Neural Networks (ONNs) related to training efficiency, nonlinear function implementation, and large input data processing. We introduce Two-Pass Forward Propagation, a novel training method that avoids specific nonlinear activation functions by modulating and re-entering error with random noise. We propose a new way to implement convolutional neural networks using simple neural networks in integrated optical systems.
  arXiv  Detail & Related papers  (2024-08-15T11:27:01Z)
• Unconventional Computing based on Four Wave Mixing in Highly Nonlinear Waveguides [0.0]
  We numerically analyze a photonic unconventional accelerator based on the four-wave mixing effect in highly nonlinear waveguides. By exploiting the rich Kerr-induced nonlinearities, multiple nonlinear transformations of an input signal can be generated and used for solving complex nonlinear tasks.
  arXiv  Detail & Related papers  (2024-02-14T12:34:38Z)
• Optical Neural Ordinary Differential Equations [44.97261923694945]
  We propose the optical neural ordinary differential equations (ON-ODE) architecture that parameterizes the continuous dynamics of hidden layers with optical ODE solvers. The ON-ODE comprises the PNNs followed by the photonic integrator and optical feedback loop, which can be configured to represent residual neural networks (ResNet) and recurrent neural networks with effectively reduced chip area occupancy.
  arXiv  Detail & Related papers  (2022-09-26T04:04:02Z)
• All-optical graph representation learning using integrated diffractive photonic computing units [51.15389025760809]
  Photonic neural networks perform brain-inspired computations using photons instead of electrons. We propose an all-optical graph representation learning architecture, termed diffractive graph neural network (DGNN) We demonstrate the use of DGNN extracted features for node and graph-level classification tasks with benchmark databases and achieve superior performance.
  arXiv  Detail & Related papers  (2022-04-23T02:29:48Z)
• Rapid characterisation of linear-optical networks via PhaseLift [51.03305009278831]
  Integrated photonics offers great phase-stability and can rely on the large scale manufacturability provided by the semiconductor industry. New devices, based on such optical circuits, hold the promise of faster and energy-efficient computations in machine learning applications. We present a novel technique to reconstruct the transfer matrix of linear optical networks.
  arXiv  Detail & Related papers  (2020-10-01T16:04:22Z)
• Limited-angle tomographic reconstruction of dense layered objects by dynamical machine learning [68.9515120904028]
  Limited-angle tomography of strongly scattering quasi-transparent objects is a challenging, highly ill-posed problem. Regularizing priors are necessary to reduce artifacts by improving the condition of such problems. We devised a recurrent neural network (RNN) architecture with a novel split-convolutional gated recurrent unit (SC-GRU) as the building block.
  arXiv  Detail & Related papers  (2020-07-21T11:48:22Z)
• Training End-to-End Analog Neural Networks with Equilibrium Propagation [64.0476282000118]
  We introduce a principled method to train end-to-end analog neural networks by gradient descent. We show mathematically that a class of analog neural networks (called nonlinear resistive networks) are energy-based models. Our work can guide the development of a new generation of ultra-fast, compact and low-power neural networks supporting on-chip learning.
  arXiv  Detail & Related papers  (2020-06-02T23:38:35Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.