The asymptotic performance of linear echo state neural networks

Abstract : In this article, a study of the mean-square error (MSE) performance of linear echo-state neural networks is performed, both for training and testing tasks. Considering the realistic setting of noise present at the network nodes, we derive deterministic equivalents for the aforementioned MSE in the limit where the number of input data T and network size n both grow large. Specializing then the network connectivity matrix to specific random settings, we further obtain simple formulas that provide new insights on the performance of such networks.
Liste complète des métadonnées

Cited literature [14 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01322809
Contributor : Hafiz Tiomoko Ali <>
Submitted on : Friday, May 27, 2016 - 5:03:09 PM
Last modification on : Thursday, February 7, 2019 - 5:48:46 PM
Document(s) archivé(s) le : Sunday, August 28, 2016 - 10:52:33 AM

File

1603.07866.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01322809, version 1

Citation

Romain Couillet, Gilles Wainrib, Harry Sevi, Hafiz Tiomoko Ali. The asymptotic performance of linear echo state neural networks. Journal of Machine Learning Research, Journal of Machine Learning Research, 2016, 17 (178), pp.1-35. ⟨hal-01322809⟩

Share

Metrics

Record views

308

Files downloads

79