Publications

Non-instantaneous information transfer in physical reservoir computing

Published |

In: Kostitsyna I., Orponen P. (eds) Unconventional Computation and Natural Computation. UCNC 2021. Lecture Notes in Computer Science, vol 12984. Springer, Cham. https://doi.org/10.1007/978-3-030-87993-8_11

| 28 Mar 2022
Author(s) |

S. Stepney

DOI |

https://doi.org/10.1007/978-3-030-87993-8_11

Abstract

The Echo State Network reservoir computing model is used in many applications. The original equations have a form of ‘instantaneous’ information flow. Here we suggest using a more physically realistic form of the equations, which takes into account the time needed for information to flow and to be processed. We demonstrate the effect of this change on a timeseries task, sunspot prediction, on a dynamical system emulation task, NARMA10, and on formulating ‘reservoir of reservoir’ equations.

DOI | https://doi.org/10.1007/978-3-030-87993-8_11