A Practical Guide to Applying Echo State Networks

By Mantas Lukosevicius. Neural Networks, 2012

862 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/11bb0941b1f6088783e26d7f9603789ee1db7faa

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

In Electrical and Electronic Engineering

Citations in the corpus

Backpropagation algorithms and Reservoir Computing in Recurrent Neural Networks for the forecasting of complex spatiotemporal dynamics (2019, background): “Reservoir Computing (RC) aims to alleviate the difficulty in learning the recurrent connections of RNNs and reduce their training time (Lukoševičius, 2012; Lukoševičius & Jaeger, 2009).”
Using Conceptors to Manage Neural Long-Term Memories for Temporal Patterns (2017, background): “The task is to store two dissimilar periodic temporal patterns in a tiny RNN and subsequently retrieve them.”
Recent Advances in Physical Reservoir Computing: A Review (2018, methodology): “Specifically, RC is a unified computational framework (Lukoševičius & Jaeger, 2009; Verstraeten, Schrauwen, d’ Haene, & Stroobandt, 2007), derived from independently proposed RNN models, such as echo state networks (ESNs) (Jaeger, 2001) and liquid state machines (LSMs) (Maass, Natschläger, & Markram”
Training Neural Networks Without Gradients: A Scalable ADMM Approach (2016, background): “Recurrent nets, which complicate standard gradient methods (Jaeger, 2002; Lukoˇseviˇcius, 2012), pose no difﬁculty for ADMM schemes whatsoever because they decouple layers using auxiliary variables.”
Machine Learning for Wireless Networks with Artificial Intelligence: A Tutorial on Neural Networks (2017, background): “3) Example RNN – Echo State Networks: Next, to shed more light on how RNNs operate, we introduce a type of RNNs that is conceptually simple and easy to implement, called echo state networks (ESNs) [92].”
Caching in the Sky: Proactive Deployment of Cache-Enabled Unmanned Aerial Vehicles for Optimized Quality-of-Experience (2016, methodology): “To ensure that the reservoir has the echo state property, Wj is defined as a sparse matrix with a spectral radius less than one [32].”
Artificial Neural Networks-Based Machine Learning for Wireless Networks: A Tutorial (2017, methodology): “The setting of other ESN components to guarantee the echo state property and to optimize ESN performance can be found in [70].”
At the Edge of Chaos: How Cerebellar Granular Layer Network Dynamics Can Provide the Basis for Temporal Filters (2015, methodology): “While teacher forcing, a strategy used in reservoir computing is able to tackle unstable learning tasks [27] by providing the target output signals before the weights are even learned, it remains to be shown if learning at Purkinje synapses itself can be stable during output feedback.”