Exploring the hidden dimensions of an optical extreme learning machine

被引:2
作者
Silva, Duarte [1 ,2 ]
Ferreira, Tiago [1 ,2 ]
Moreira, Felipe C. [1 ,2 ]
Rosa, Carla C. [1 ,2 ]
Guerreiro, Ariel [1 ,2 ]
Silva, Nuno A. [1 ,2 ]
机构
[1] Univ Porto, Fac Ciencias, Dept Fis & Astron, Rua Campo Alegre S-N, P-4169007 Porto, Portugal
[2] INESC TEC, Ctr Appl Photon, Rua Campo Alegre 687, P-4169007 Porto, Portugal
关键词
Extreme Learning Machine; Optical Computing; Machine Learning; Optics;
D O I
10.1051/jeos/2023001
中图分类号
O43 [光学];
学科分类号
070207 ; 0803 ;
摘要
Extreme Learning Machines (ELMs) are a versatile Machine Learning (ML) algorithm that features as the main advantage the possibility of a seamless implementation with physical systems. Yet, despite the success of the physical implementations of ELMs, there is still a lack of fundamental understanding in regard to their optical implementations. In this context, this work makes use of an optical complex media and wavefront shaping techniques to implement a versatile optical ELM playground to gain a deeper insight into these machines. In particular, we present experimental evidences on the correlation between the effective dimensionality of the hidden space and its generalization capability, thus bringing the inner workings of optical ELMs under a new light and opening paths toward future technological implementations of similar principles.
引用
收藏
页码:436 / 444
页数:4
相关论文
共 15 条
  • [1] A Deep Learning Approach to Ancient Egyptian Hieroglyphs Classification
    Barucci, Andrea
    Cucci, Costanza
    Franci, Massimiliano
    Loschiavo, Marco
    Argenti, Fabrizio
    [J]. IEEE ACCESS, 2021, 9 : 123438 - 123447
  • [2] Deng L.., 2012, IEEE signal processing magazine, V29
  • [3] Goodman J. W., 2020, Speckle phenomena in optics: theory and applications, V2nd
  • [4] HORN R. A., 2012, Matrix Analysis
  • [5] Extreme Learning Machine for Regression and Multiclass Classification
    Huang, Guang-Bin
    Zhou, Hongming
    Ding, Xiaojian
    Zhang, Rui
    [J]. IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 2012, 42 (02): : 513 - 529
  • [6] BINARY COMPUTER-GENERATED HOLOGRAMS
    LEE, WH
    [J]. APPLIED OPTICS, 1979, 18 (21): : 3661 - 3669
  • [7] Theory of Neuromorphic Computing by Waves: Machine Learning by Rogue Waves, Dispersive Shocks, and Solitons
    Marcucci, Giulia
    Pierangeli, Davide
    Conti, Claudio
    [J]. PHYSICAL REVIEW LETTERS, 2020, 125 (09)
  • [8] Deep learning
    Rusk, Nicole
    [J]. NATURE METHODS, 2016, 13 (01) : 35 - 35
  • [9] Saade A, 2016, INT CONF ACOUST SPEE, P6215, DOI 10.1109/ICASSP.2016.7472872
  • [10] Schrauwen B., 2007, PROC 15 EUR SYMPSOSI, P471