Ghent University Academic Bibliography

Advanced

Pruning and regularization in Reservoir Computing: a first insight

Xavier Dutoit, Benjamin Schrauwen UGent, Jan Van Campenhout UGent, Dirk Stroobandt UGent, Hendrik Van Brusssel and Marnix Nuttin (2008) European Symposium on Artificial Neural Networks, 16th, Proceedings.
abstract
Reservoir Computing is a new paradigm for using Recurrent Neural Networks which shows promising results. However, as the recurrent part is created randomly, it typically needs to be large enough to be able to capture the dynamic features of the data considered. Moreover, this random creation is still lacking a strong methodology. We propose to study how pruning some connections from the reservoir to the readout can help on the one hand to increase the generalisation ability, in much the same way as regularisation techniques do, and on the other hand to improve the implementability of reservoirs in hardware. Furthermore we study the actual sub-reservoir which is kept after pruning which leads to important insights in what we have to expect from a good reservoir.
Please use this url to cite or link to this publication:
author
organization
year
type
conference
publication status
published
subject
keyword
pruning, reservoir computing, regularization
in
European Symposium on Artificial Neural Networks, 16th, Proceedings
pages
6 pages
publisher
d-side publications
conference name
16th European symposium on Artificial Neural Networks
conference location
Brugge, Belgium
conference start
2008-04-23
conference end
2008-04-25
language
English
UGent publication?
yes
classification
C1
id
680771
handle
http://hdl.handle.net/1854/LU-680771
date created
2009-06-05 15:38:31
date last changed
2010-02-25 17:20:31
@inproceedings{680771,
  abstract     = {Reservoir Computing is a new paradigm for using Recurrent Neural Networks which shows promising results. However, as the recurrent part is created randomly, it typically needs to be large enough to be able to capture the dynamic features of the data considered. Moreover, this random creation is still lacking a strong methodology. We propose to study how pruning some connections from the reservoir to the readout can help on the one hand to increase the generalisation ability, in much the same way as regularisation techniques do, and on the other hand to improve the implementability of reservoirs in hardware. Furthermore we study the actual sub-reservoir which is kept after pruning which leads to important insights in what we have to expect from a good reservoir.},
  author       = {Dutoit, Xavier and Schrauwen, Benjamin and Van Campenhout, Jan and Stroobandt, Dirk and Van Brusssel, Hendrik and Nuttin, Marnix},
  booktitle    = {European Symposium on Artificial Neural Networks, 16th, Proceedings},
  keyword      = {pruning,reservoir computing,regularization},
  language     = {eng},
  location     = {Brugge, Belgium},
  pages        = {6},
  publisher    = {d-side publications},
  title        = {Pruning and regularization in Reservoir Computing: a first insight},
  year         = {2008},
}

Chicago
Dutoit, Xavier, Benjamin Schrauwen, Jan Van Campenhout, Dirk Stroobandt, Hendrik Van Brusssel, and Marnix Nuttin. 2008. “Pruning and Regularization in Reservoir Computing: a First Insight.” In European Symposium on Artificial Neural Networks, 16th, Proceedings. d-side publications.
APA
Dutoit, X., Schrauwen, B., Van Campenhout, J., Stroobandt, D., Van Brusssel, H., & Nuttin, M. (2008). Pruning and regularization in Reservoir Computing: a first insight. European Symposium on Artificial Neural Networks, 16th, Proceedings. Presented at the 16th European symposium on Artificial Neural Networks, d-side publications.
Vancouver
1.
Dutoit X, Schrauwen B, Van Campenhout J, Stroobandt D, Van Brusssel H, Nuttin M. Pruning and regularization in Reservoir Computing: a first insight. European Symposium on Artificial Neural Networks, 16th, Proceedings. d-side publications; 2008.
MLA
Dutoit, Xavier, Benjamin Schrauwen, Jan Van Campenhout, et al. “Pruning and Regularization in Reservoir Computing: a First Insight.” European Symposium on Artificial Neural Networks, 16th, Proceedings. d-side publications, 2008. Print.