Reservoir computing
Reservoir computing is a framework for computation derived from recurrent neural network theory that maps input signals into higher dimensional computational spaces through the dynamics of a fixed, non-linear system called a reservoir. After the input signal is fed into the reservoir, which is treated as a "black box," a simple readout mechanism is trained to read the state of the reservoir and map it to the desired output
Resources
Echo state networks (ESN)
- https://en.wikipedia.org/wiki/Echo_state_network
- The ESN is a recurrent neural network with a sparsely connected hidden layer (with typically 1% connectivity)
References
Echo state networks (ESN)
-
The ESN is a recurrent neural network with a sparsely connected hidden layer (with typically 1% connectivity)
-
#PAPER Harnessing nonlinearity: predicting chaotic systems and saving energy in wireless communication (Jaeger 2004)
-
#PAPER Using Machine Learning to Replicate Chaotic Attractors and Calculate Lyapunov Exponents from Data (Pathak 2017)
-
#PAPER Model-Free Prediction of Large Spatiotemporally Chaotic Systems from Data: A Reservoir Computing Approach (Pathak 2018)
-
#PAPER Wind Power Forecasting Based on Echo State Networks and Long Short-Term Memory (Lopez 2018)
- ESN + LSTM
-
#PAPER Comparison between DeepESNs and gatedRNNs on multivariate time-series prediction (Gallicchio 2019)
-
#PAPER Deep Echo State Network (DeepESN): A Brief Survey (Gallicchio 2020)
-
#PAPER Comparison of Recurrent Neural Networks for Wind Power Forecasting (Lopez 2020)
- ESN + LSTM