Academic Bibliography

 Search 200 years of publications by Ghent University researchers.
Advanced search
1 file | 1.78 MB
Add to list
  1. Search results
  2. Training passive photonic reservoirs ...

Training passive photonic reservoirs with integrated optical readout

Matthias Freiberger (UGent) , Andrew Katumba (UGent) , Peter Bienstman (UGent) and Joni Dambre (UGent)
(2019) IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS. 30(7). p.1943-1953
Author
Organization
Project
  • PHRESCO (PHRESCO: PHotonic REservoir COmputing)
Abstract
As Moore's law comes to an end, neuromorphic approaches to computing are on the rise. One of these, passive photonic reservoir computing, is a strong candidate for computing at high bitrates (>10 Gb/s) and with low energy consumption. Currently though, both benefits are limited by the necessity to perform training and readout operations in the electrical domain. Thus, efforts are currently underway in the photonic community to design an integrated optical readout, which allows to perform all operations in the optical domain. In addition to the technological challenge of designing such a readout, new algorithms have to be designed in order to train it. Foremost, suitable algorithms need to be able to deal with the fact that the actual on-chip reservoir states are not directly observable. In this paper, we investigate several options for such a training algorithm and propose a solution in which the complex states of the reservoir can be observed by appropriately setting the readout weights, while iterating over a predefined input sequence. We perform numerical simulations in order to compare our method with an ideal baseline requiring full observability as well as with an established black-box optimization approach (CMA-ES).
Keywords
SYSTEMS, Cognitive computing, integrated optical readout, limited observability, neuromorphic computing, nonlinearity inversion, photonic computing, reservoir computing

Downloads

  • Download
    (...).pdf
    • full text (Published version)
    • |
    • UGent only
    • |
    • PDF
    • |
    • 1.78 MB

Citation

Please use this url to cite or link to this publication:

MLA
Freiberger, Matthias, et al. “Training Passive Photonic Reservoirs with Integrated Optical Readout.” IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, vol. 30, no. 7, 2019, pp. 1943–53, doi:10.1109/TNNLS.2018.2874571.
APA
Freiberger, M., Katumba, A., Bienstman, P., & Dambre, J. (2019). Training passive photonic reservoirs with integrated optical readout. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, 30(7), 1943–1953. https://doi.org/10.1109/TNNLS.2018.2874571
Chicago author-date
Freiberger, Matthias, Andrew Katumba, Peter Bienstman, and Joni Dambre. 2019. “Training Passive Photonic Reservoirs with Integrated Optical Readout.” IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 30 (7): 1943–53. https://doi.org/10.1109/TNNLS.2018.2874571.
Chicago author-date (all authors)
Freiberger, Matthias, Andrew Katumba, Peter Bienstman, and Joni Dambre. 2019. “Training Passive Photonic Reservoirs with Integrated Optical Readout.” IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 30 (7): 1943–1953. doi:10.1109/TNNLS.2018.2874571.
Vancouver
1.
Freiberger M, Katumba A, Bienstman P, Dambre J. Training passive photonic reservoirs with integrated optical readout. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS. 2019;30(7):1943–53.
IEEE
[1]
M. Freiberger, A. Katumba, P. Bienstman, and J. Dambre, “Training passive photonic reservoirs with integrated optical readout,” IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS, vol. 30, no. 7, pp. 1943–1953, 2019.
@article{8627132,
  abstract     = {{As Moore's law comes to an end, neuromorphic approaches to computing are on the rise. One of these, passive photonic reservoir computing, is a strong candidate for computing at high bitrates (>10 Gb/s) and with low energy consumption. Currently though, both benefits are limited by the necessity to perform training and readout operations in the electrical domain. Thus, efforts are currently underway in the photonic community to design an integrated optical readout, which allows to perform all operations in the optical domain. In addition to the technological challenge of designing such a readout, new algorithms have to be designed in order to train it. Foremost, suitable algorithms need to be able to deal with the fact that the actual on-chip reservoir states are not directly observable. In this paper, we investigate several options for such a training algorithm and propose a solution in which the complex states of the reservoir can be observed by appropriately setting the readout weights, while iterating over a predefined input sequence. We perform numerical simulations in order to compare our method with an ideal baseline requiring full observability as well as with an established black-box optimization approach (CMA-ES).}},
  author       = {{Freiberger, Matthias and Katumba, Andrew and Bienstman, Peter and Dambre, Joni}},
  issn         = {{2162-237X}},
  journal      = {{IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS}},
  keywords     = {{SYSTEMS,Cognitive computing,integrated optical readout,limited observability,neuromorphic computing,nonlinearity inversion,photonic computing,reservoir computing}},
  language     = {{eng}},
  number       = {{7}},
  pages        = {{1943--1953}},
  title        = {{Training passive photonic reservoirs with integrated optical readout}},
  url          = {{http://dx.doi.org/10.1109/TNNLS.2018.2874571}},
  volume       = {{30}},
  year         = {{2019}},
}
Altmetric
View in Altmetric
Web of Science
Times cited: