Matthew N. O. Sadiku, Kelechi G. Eze, Sarhan M. Musa
Reservoir computing (RC) is a new computing paradigm that allows harnessing the dynamics of a reservoir (or compute core) to perform computations. RC is a new training concept for recurrent neural networks. It initially emerged as a software-only technique and used as an algorithmic way of processing temporal data. It has been successfully used in pattern classification problems such as like speech and image recognition and time series prediction. This paper provides a brief primer to basic concepts, implementations, and applications of reservoir computing
Reservoir Computing, Photonic Reservoir Computing, Recurrent Neural Networks