Academic Bibliography

 Search 200 years of publications by Ghent University researchers.
Advanced search
1 file | 829.17 KB
Add to list
  1. Search results
  2. Trainable hardware for dynamical comp...

Trainable hardware for dynamical computing using error backpropagation through physical media

Michiel Hermans (UGent) , Michaël Burm (UGent) , Thomas Van Vaerenbergh (UGent) , Joni Dambre (UGent) and Peter Bienstman (UGent)
(2015) NATURE COMMUNICATIONS. 6.
Author
Organization
Abstract
Neural networks are currently implemented on digital Von Neumann machines, which do not fully leverage their intrinsic parallelism. We demonstrate how to use a novel class of reconfigurable dynamical systems for analogue information processing, mitigating this problem. Our generic hardware platform for dynamic, analogue computing consists of a reciprocal linear dynamical system with nonlinear feedback. Thanks to reciprocity, a ubiquitous property of many physical phenomena like the propagation of light and sound, the error backpropagation-a crucial step for tuning such systems towards a specific task-can happen in hardware. This can potentially speed up the optimization process significantly, offering important benefits for the scalability of neuro-inspired hardware. In this paper, we show, using one experimentally validated and one conceptual example, that such systems may provide a straightforward mechanism for constructing highly scalable, fully dynamical analogue computers.
Keywords
NETWORKS, backpropagation. RESERVOIR, PARALLEL, STATES, reservoir computing, photonics, accoustics, analog computing, machine learning

Downloads

  • Download
    ncomms7729.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 829.17 KB

Citation

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

MLA
Hermans, Michiel, et al. “Trainable Hardware for Dynamical Computing Using Error Backpropagation through Physical Media.” NATURE COMMUNICATIONS, vol. 6, 2015, doi:10.1038/ncomms7729.
APA
Hermans, M., Burm, M., Van Vaerenbergh, T., Dambre, J., & Bienstman, P. (2015). Trainable hardware for dynamical computing using error backpropagation through physical media. NATURE COMMUNICATIONS, 6. https://doi.org/10.1038/ncomms7729
Chicago author-date
Hermans, Michiel, Michaël Burm, Thomas Van Vaerenbergh, Joni Dambre, and Peter Bienstman. 2015. “Trainable Hardware for Dynamical Computing Using Error Backpropagation through Physical Media.” NATURE COMMUNICATIONS 6. https://doi.org/10.1038/ncomms7729.
Chicago author-date (all authors)
Hermans, Michiel, Michaël Burm, Thomas Van Vaerenbergh, Joni Dambre, and Peter Bienstman. 2015. “Trainable Hardware for Dynamical Computing Using Error Backpropagation through Physical Media.” NATURE COMMUNICATIONS 6. doi:10.1038/ncomms7729.
Vancouver
1.
Hermans M, Burm M, Van Vaerenbergh T, Dambre J, Bienstman P. Trainable hardware for dynamical computing using error backpropagation through physical media. NATURE COMMUNICATIONS. 2015;6.
IEEE
[1]
M. Hermans, M. Burm, T. Van Vaerenbergh, J. Dambre, and P. Bienstman, “Trainable hardware for dynamical computing using error backpropagation through physical media,” NATURE COMMUNICATIONS, vol. 6, 2015.
@article{5912189,
  abstract     = {{Neural networks are currently implemented on digital Von Neumann machines, which do not fully leverage their intrinsic parallelism. We demonstrate how to use a novel class of reconfigurable dynamical systems for analogue information processing, mitigating this problem. Our generic hardware platform for dynamic, analogue computing consists of a reciprocal linear dynamical system with nonlinear feedback. Thanks to reciprocity, a ubiquitous property of many physical phenomena like the propagation of light and sound, the error backpropagation-a crucial step for tuning such systems towards a specific task-can happen in hardware. This can potentially speed up the optimization process significantly, offering important benefits for the scalability of neuro-inspired hardware. In this paper, we show, using one experimentally validated and one conceptual example, that such systems may provide a straightforward mechanism for constructing highly scalable, fully dynamical analogue computers.}},
  articleno    = {{6729}},
  author       = {{Hermans, Michiel and Burm, Michaël and Van Vaerenbergh, Thomas and Dambre, Joni and Bienstman, Peter}},
  issn         = {{2041-1723}},
  journal      = {{NATURE COMMUNICATIONS}},
  keywords     = {{NETWORKS,backpropagation. RESERVOIR,PARALLEL,STATES,reservoir computing,photonics,accoustics,analog computing,machine learning}},
  language     = {{eng}},
  pages        = {{8}},
  title        = {{Trainable hardware for dynamical computing using error backpropagation through physical media}},
  url          = {{http://doi.org/10.1038/ncomms7729}},
  volume       = {{6}},
  year         = {{2015}},
}
Altmetric
View in Altmetric
Web of Science
Times cited: