A Random Matrix Approach to Echo-State Neural Networks

Abstract : Recurrent neural networks, especially in their linear version, have provided many qualitative insights on their performance under different configurations. This article provides, through a novel random matrix framework, the quantitative counterpart of these performance results, particularly in the case of echo-state networks. Beyond mere insights, our approach conveys a deeper understanding on the core mechanism under play for both training and testing.
Complete list of metadatas

Cited literature [14 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-01812026
Contributor : Hafiz Tiomoko Ali <>
Submitted on : Monday, July 9, 2018 - 4:45:20 PM
Last modification on : Monday, December 17, 2018 - 2:28:11 PM
Long-term archiving on : Monday, October 1, 2018 - 2:40:37 AM

File

couillet16.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-01812026, version 1

Citation

Romain Couillet, Gilles Wainrib, Hafiz Tiomoko Ali, Harry Sevi. A Random Matrix Approach to Echo-State Neural Networks. International Conference on Machine Learning (ICML), New York, USA, 2016, Jun 2016, New York, United States. ⟨hal-01812026⟩

Share

Metrics

Record views

240

Files downloads

33