In reservoir computing the readout is a relatively simple machine learning stage such as a neural network or linear classifier. The readout transforms the raw output of the reservoir (biological, physical or digital) into the final output.. As the raw output is effectively some sort of rich non-linear function of the inputs, the readout can afford to be relatively simple and easy to train.
Used in Chap. 7: page 99
Also known as readout
Reservoir computing – main stages