Fully forward mode training for optical neural networks
Open accessThe DOI matches its Crossref record; 15 of the 15 references it cites were found in a registry or an index.
Abstract
Optical computing promises to improve the speed and energy efficiency of machine learning applications 1–6. However, current approaches to efficiently train these models are limited by in silico emulation on digital computers. Here we develop a method called fully forward mode (FFM) learning, which implements the compute-intensive training process on the physical system. The majority of the machine learning operations are thus efficiently conducted in parallel on site, alleviating numerical modelling constraints. In free-space and integrated photonics, we experimentally demonstrate optical systems with state-of-the-art performances for a given network size. FFM learning shows training the deepest optical neural networks with millions of parameters achieves accuracy equivalent to the ideal model. It supports all-optical focusing through scattering media with a resolution of the diffraction limit; it can also image in parallel the objects hidden outside the direct line of sight at over a kilohertz frame rate and can conduct all-optical processing with light intensity as weak as subphoton per pixel (5.40 × 10 18- operations-per-second-per-watt energy efficiency) at room temperature. Furthermore, we prove that FFM learning can automatically search non-Hermitian exceptional points without an analytical model. FFM learning not only facilitates orders-of-magnitude-faster learning processes, but can also advance applied and theoretical fields such as deep neural networks, ultrasensitive perception and topological photonics.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
LeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Gradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805
Machine learning: Trends, perspectives, and prospects
Jordan, M. I., & Mitchell, T. M. (2015) · Science · doi:10.1126/science.aaa8415 · Cited by 9,955 · cited 2×
“Optical computing promises to improve the speed and energy efficiency of machine learning applications 1–6 .”
BackgroundFashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
Xiao, H., Rasul, K., & Vollgraf, R. (2017) · arXiv · doi:10.48550/arxiv.1708.07747 · Cited by 6,108
Lu, L., Joannopoulos, J. D., & Soljačić, M. (2014) · Nature Photonics · doi:10.1038/nphoton.2014.248 · Cited by 3,655 · cited 1×
“Previous studies needed to first model the system offline and care-fully approach the theoretical results on site by carefully tuning the gain and loss 56,65–67 .”
BackgroundDeep learning with coherent nanophotonic circuits
Shen, Y., Harris, N. C., Skirlo, S., Prabhu, M., Baehr-Jones, T., Hochberg, M., Sun, X., Zhao, S., Larochelle, H., Englund, D., et al. (2017) · Nature Photonics · doi:10.1038/nphoton.2017.93 · Cited by 3,340
The chips are down for Moore’s law
Waldrop, M. M. (2016) · Nature · doi:10.1038/530144a · Cited by 2,397
Watson, J. L., Juergens, D., Bennett, N. R., Trippe, B. L., Yim, J., Eisenach, H. E., Ahern, W., Borst, A. J., Ragotte, R. J., Milles, L. F., et al. (2022) · doi:10.1101/2022.12.09.519842 · Cited by 2,369
De novo design of protein structure and function with RFdiffusion
Watson, J. L., Juergens, D., Bennett, N. R., Trippe, B. L., Yim, J., Eisenach, H. E., Ahern, W., Borst, A. J., Ragotte, R. J., Milles, L. F., et al. (2023) · Nature · doi:10.1038/s41586-023-06415-8 · Cited by 2,325
Parallel convolutional processing using an integrated photonic tensor core
Feldmann, J., Youngblood, N., Karpov, M., Gehring, H., Li, X., Stappers, M., Le Gallo, M., Fu, X., Lukashchuk, A., Raja, A. S., et al. (2021) · Nature · doi:10.1038/s41586-020-03070-1 · Cited by 1,900
Inverse design in nanophotonics
Molesky, S., Lin, Z., Piggott, A. Y., Jin, W., Vucković, J., & Rodriguez, A. W. (2018) · Nature Photonics · doi:10.1038/s41566-018-0246-9 · Cited by 1,818
Controlling waves in space and time for imaging and focusing in complex media
Mosk, A. P., Lagendijk, A., Lerosey, G., & Fink, M. (2012) · Nature Photonics · doi:10.1038/nphoton.2012.88 · Cited by 1,536 · cited 1×
“Focusing through scattering media with high precision has applications in microscopic as well as macroscopic imaging 59,60 .”
MethodInverse design and demonstration of a compact and broadband on-chip wavelength demultiplexer
Piggott, A. Y., Lu, J., Lagoudakis, K. G., Petykiewicz, J., Babinec, T. M., & Vučković, J. (2015) · Nature Photonics · doi:10.1038/nphoton.2015.69 · Cited by 1,113
Magnetic control of tokamak plasmas through deep reinforcement learning
Degrave, J., Felici, F., Buchli, J., Neunert, M., Tracey, B., Carpanese, F., Ewalds, T., Hafner, R., Abdolmaleki, A., de las Casas, D., et al. (2022) · Nature · doi:10.1038/s41586-021-04301-9 · Cited by 751 · cited 18×
“In addition to the compute-intensive data and error propagation, the loss and gradient calculations can also be carried out through onsite optical and electronic processing (Supplementary Note 13).”
BackgroundInfluentialAn on-chip photonic deep neural network for image classification
Ashtiani, F., Geers, A. J., & Aflatouni, F. (2022) · Nature · doi:10.1038/s41586-022-04714-0 · Cited by 699
Cited by
A guidance to intelligent metamaterials and metamaterials intelligence
Qian, C., Kaminer, I., & Chen, H. (2025) · Nature Communications · doi:10.1038/s41467-025-56122-3 · Cited by 254
Metadata
- Authors
- Xue, Z., Zhou, T., Xu, Z., Yu, S., Dai, Q., & Fang, L.
- Venue
- Nature
- Published
- 2024-08-07
- Volume
- 632
- Issue
- 8024
- Pages
- 280-286
- ISSN
- 0028-0836 (ISSN-L)
- 1476-4687
- OpenAlex
- W4401389039
- Cited by
- 165
- References
- 15
- FWCI
- 24.56
- Type
- journal article
- Language
- en
- Open access
- hybrid
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.