Photonics Research, Volume. 10, Issue 12, 2846(2022)

Large-scale photonic natural language processing

Carlo M. Valensise1, Ivana Grecco2, Davide Pierangeli1,2,3、*, and Claudio Conti1,2,3
Author Affiliations
  • 1Enrico Fermi Research Center (CREF), 00184 Rome, Italy
  • 2Physics Department, Sapienza University of Rome, 00185 Rome, Italy
  • 3Institute for Complex Systems, National Research Council (ISC-CNR), 00185 Rome, Italy
  • show less
    Figures & Tables(5)
    Three-dimensional PELM for language processing. (A) The text database entry is a paragraph of variable length. Text pre-processing: a sparse representation of the input paragraph is mapped into a Hadamard matrix with phase values in [0,π]. (B) The mask is encoded into the optical wavefront by a phase-only SLM. Free-space propagation of the optical field maps the input data into a 3D intensity distribution (speckle-like volume). (C) Sampling the propagating laser beam in multiple far-field planes enables upscaling the feature space. Intensities picked from all the spatial modes form the output layer H3D that undergoes training via ridge regression. By using three planes (j=3), we get a network capacity C>1010. (D) The example shows a binary text classification problem for large-scale rating.
    Photonic sentiment analysis. (A), (B) Training and test accuracy of the 3D-PELM on the IMDb dataset as a function of the number of output channels. The shaded area corresponds to the over-parameterized region. The configuration in (B) allows us to reach very high accuracy in the over-parameterized region with a dataset limited to Ntrain=1186 training points. In (A), the same accuracy is reached in the under-parameterized region with Ntrain=12,278. Black horizontal lines correspond to the maximum test accuracy achieved (0.77). (C) IMDb classification accuracy by varying the number of features M and training dataset size Ntrain. The boundary between the under and over-parameterized region (interpolation threshold), Ntrain=M, is characterized by a sharp accuracy drop (cyan contour line).
    Performances at ultralarge scale. (A)–(C) Test accuracy as a function of M for different input sizes L. In all cases, the 3D-PELM performance saturates in the over-parameterized region, reaching a plateau. A linear fit of the data preceding the plateau shows that the onset of the saturation is faster for datasets with a larger input space. The corresponding angular coefficient m is inset in each panel. (D) Test accuracy varying the training set size for M=0.8×105 and M=1.2×105.
    Analysis of the IMDb accuracy. (A), (B) The comparison reports the accuracy for the experimental device (3D-PELM device), the simulated device (3D-PELM numerics), the random projection method with ridge regression (RP), the support vector machine (SVM), and a convolutional neural network (CNN) in both the under-parameterized (M=1×103) and over-parameterized (M=4×104) regimes, for (A) Ntrain=6700 and (B) Ntrain=1500. 8-bit numerical results, when applicable, refer to the over-parameterized regime.
    • Table 1. Maximum Network Capacity of Current Photonic Neuromorphic Computing Hardware for Supervised Learning

      View table
      View in Article

      Table 1. Maximum Network Capacity of Current Photonic Neuromorphic Computing Hardware for Supervised Learning

      Working PrincipleMLCMachine Learning TaskRef.
      Time-multiplexed cavity14007129107Regression[39]
      Amplitude modulation16,3842000108Human action recognition[27]
      Frequency multiplexing200640105Time series recovery[41]
      Optical multiple scattering50,00064106Chaotic series prediction[38]
      Amplitude Fourier filtering102443,263107Image classification[30]
      Multimode fiber240240105Classification, regression[35]
      Free-space propagation6400784106Classification, regression[34]
      3D optical field120,000131,0441010Natural language processing3D-PELM

    Get Citation

    Copy Citation Text

    Carlo M. Valensise, Ivana Grecco, Davide Pierangeli, Claudio Conti. Large-scale photonic natural language processing[J]. Photonics Research, 2022, 10(12): 2846

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: Optical Devices

    Received: Aug. 10, 2022

    Accepted: Oct. 8, 2022

    Published Online: Nov. 24, 2022

    The Author Email: Davide Pierangeli (