Echo state networks are universal
By Lyudmila Grigoryeva, Juan-Pablo Ortega. Neural Networks, 2018
301 citations recorded by the source corpus.
This paper shows that echo state networks are universal uniform approximants in the context of discrete-time fading memory filters with uniformly bounded inputs defined on negative infinite times. This result guarantees that any fading memory input/output system in discrete time can be realized as a simple finite-dimensional neural network-type state-space model with a static linear readout map. This approximation is valid for infinite time intervals. The proof of this statement is based on fundamental results, also presented in this work, about the topological nature of the fading memory property and about reservoir computing systems generated by continuous reservoir maps.
Subfields: Artificial Intelligence, Electrical and Electronic Engineering
View Paper: https://www.semanticscholar.org/paper/65926aa45da388a1f36efb473bd517a0bdd97226
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Petros Koumoutsakos
- Themistoklis P. Sapsis
- Pantelis R. Vlachas
- Michelle Girvan
- Jaideep Pathak
- Edward Ott
- Brian R. Hunt
- Seiji Takeda
In Electrical and Electronic Engineering
- Edward Ott
- Seiji Takeda
- Petros Koumoutsakos
- Brian R. Hunt
- Gouhei Tanaka
- Naoki Kanazawa
- Daiju Nakano
- Toshiyuki Yamane
Citations in the corpus
- Recent Advances in Physical Reservoir Computing: A Review (2018, background): “Recently, it was theoretically shown that ESNs have universal approximation ability in terms of discrete-time fading memory filters with uniformly bounded inputs defined on negative infinite times (Grigoryeva & Ortega, 2018).”
- Backpropagation algorithms and Reservoir Computing in Recurrent Neural Networks for the forecasting of complex spatiotemporal dynamics (2019, background): “This success has sparked the interest of theoretical researchers that proved universal approximation properties of these models (Gonon & Ortega, 2019; Grigoryeva & Ortega, 2018).”