mlsanityData to 7 Oct 2026

Training of photonic neural networks through in situ backpropagation

arXiv 1805.09943 ↗First posted 25 May 2018physics.optics cs.LG cs.NEcited by 45

Not on the map. Why?

Abstract (arXiv)

Recently, integrated optics has gained interest as a hardware platform for implementing machine learning algorithms. Of particular interest are artificial neural networks, since matrix-vector multi- plications, which are used heavily in artificial neural networks, can be done efficiently in photonic circuits. The training of an artificial neural network is a crucial step in its application. However, currently on the integrated photonics platform there is no efficient protocol for the training of these networks. In this work, we introduce a method that enables highly efficient, in situ training of a photonic neural network. We use adjoint variable methods to derive the photonic analogue of the backpropagation algorithm, which is the standard method for computing gradients of conventional neural networks. We further show how these gradients may be obtained exactly by performing intensity measurements within the device. As an application, we demonstrate the training of a numerically simulated photonic artificial neural network. Beyond the training of photonic machine learning implementations, our method may also be of broad interest to experimental sensitivity analysis of photonic systems and the optimization of reconfigurable optics platforms.

Key facts. Cited by 45 papers in the map (data through 7 Oct 2026). First posted on arXiv 25 May 2018.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 7; 2022: 6; 2023: 8; 2024: 9; 2025: 8; 2026: 7.

45 papers.

Sort:

  1. 2021Deep physical neural networks enabled by a backpropagation algorithm for arbitrary physical systemscited by 125
  2. 2024Training of Physical Neural Networkscited by 47
  3. 2022Experimentally realized in situ backpropagation for deep learning in nanophotonic neural networkscited by 43
  4. 2023Backpropagation-free Training of Deep Physical Neural Networkscited by 42
  5. 2021Self-learning Machines based on Hamiltonian Echo Backpropagationcited by 28
  6. 2021L2ight: Enabling On-Chip Learning for Optical Neural Networks via Efficient in-situ Subspace Optimizationcited by 21
  7. 2021Silicon Photonic Architecture for Training Deep Neural Networks with Direct Feedback Alignmentcited by 17
  8. 2023Nonlinear optical encoding enabled by recurrent linear scatteringcited by 14
  9. 2021A compact butterfly-style silicon photonic-electronic neural chip for hardware-efficient deep learningcited by 13
  10. 2024Quantum Equilibrium Propagation for efficient training of quantum systems based on Onsager reciprocitycited by 9
  11. 2022Dual adaptive training of photonic neural networkscited by 8
  12. 2023Training neural networks with end-to-end optical backpropagationcited by 7
33 more
  1. 2021Photonic Differential Privacy with Direct Feedback Alignmentcited by 5
  2. 2021BPLight-CNN: A Photonics-based Backpropagation Accelerator for Deep Learningcited by 4
  3. 2023Integrated multi-operand optical neurons for scalable and hardware-efficient deep learningcited by 4
  4. 2023Real-Time FJ/MAC PDE Solvers via Tensorized, Back-Propagation-Free Optical PINN Trainingcited by 4
  5. 2024Training all-mechanical neural networks for task learning through in situ backpropagationcited by 4
  6. 2022FatNet: High Resolution Kernels for Classification Using Fully Convolutional Optical Neural Networkscited by 3
  7. 2023Digital Modeling on Large Kernel Metamaterial Neural Networkcited by 3
  8. 2024Asymmetrical estimator for training encapsulated deep photonic neural networkscited by 3
  9. 2023Mirage: An RNS-Based Photonic Accelerator for DNN Trainingcited by 2
  10. 2024ExtremeMETA: High-speed Lightweight Image Segmentation Model by Remodeling Multi-channel Metamaterial Imagerscited by 2
  11. 2025Scalable Back-Propagation-Free Training of Optical Physics-Informed Neural Networkscited by 2
  12. 2025Model-free Optical Processors using In Situ Reinforcement Learning with Proximal Policy Optimizationcited by 2
  13. 2024Integration of Programmable Diffraction with Digital Neural Networkscited by 1
  14. 2025Nonlinear Computation with Linear Optics via Source-Position Encodingcited by 1
  15. 2025Circuit realization and hardware linearization of monotone operator equilibrium networkscited by 1
  16. 2026Backpropagation-Free Trunk Training via the Split Forward Gradientscited by 1
  17. 2022Physical Deep Learning with Biologically Plausible Training Methodcited by 0
  18. 2022Adversarial attacks on an optical neural networkcited by 0
  19. 2022Misaligned orientations of 4f optical neural network for image classification accuracy on various datasetscited by 0
  20. 2023Optimal Data Generation in Multi-Dimensional Parameter Spaces, using Bayesian Optimizationcited by 0
  21. 2024Thermal Crosstalk Modelling and Compensation Methods for Programmable Photonic Integrated Circuitscited by 0
  22. 2024Optical Computing for Deep Neural Network Acceleration: Foundations, Recent Developments, and Emerging Directionscited by 0
  23. 2024Training Large-Scale Optical Neural Networks with Two-Pass Forward Propagationcited by 0
  24. 2025ConvShareViT: Enhancing Vision Transformers with Convolutional Attention Mechanisms for Free-Space Optical Acceleratorscited by 0
  25. 2025Gradients of unitary optical neural networks using parameter-shift rulecited by 0
  26. 2025Spatially Parallel All-optical Neural Networkscited by 0
  27. 2025LuxIA: A Lightweight Unitary matriX-based Framework Built on an Iterative Algorithm for Photonic Neural Network Trainingcited by 0
  28. 2026Laser interferometry as a robust neuromorphic platform for machine learningcited by 0
  29. 2026All-Optical Segmentation via Diffractive Neural Networks for Autonomous Drivingcited by 0
  30. 2026Equilibrium Propagation and Hamiltonian Inference in the Diffusive Fitzhugh-Nagumo Modelcited by 0
  31. 2026Optimizing Energy-based Neural Network Training with Coherent Ising Machinecited by 0
  32. 2026Physical policy gradient theorem for in situ stochastic-adjoint trainingcited by 0
  33. 2026LCAP: Population-Informed Latent Chip Adaptation from Few Output Probes for Photonic Neural Networkscited by 0