A Substrate-Independent Framework to Characterise Reservoir Computers

TL;DR

This paper introduces a versatile framework to evaluate and compare various physical substrates for reservoir computing, enhancing understanding of their computational capabilities and guiding future substrate design.

Contribution

It presents a substrate-independent experimental framework to characterize reservoir quality, linking properties to task performance and aiding in the design of better reservoirs.

Findings

Framework effectively compares different substrates

Reconfiguration impacts reservoir quality and performance

Mapping properties to tasks improves substrate design

Abstract

The Reservoir Computing (RC) framework states that any non-linear, input-driven dynamical system (the reservoir) exhibiting properties such as a fading memory and input separability can be trained to perform computational tasks. This broad inclusion of systems has led to many new physical substrates for RC. Properties essential for reservoirs to compute are tuned through reconfiguration of the substrate, such as change in virtual topology or physical morphology. As a result, each substrate possesses a unique `quality' -- obtained through reconfiguration -- to realise different reservoirs for different tasks. Here we describe an experimental framework to characterise the quality of potentially any substrate for RC. Our framework reveals that a definition of quality is not only useful to compare substrates, but can help map the non-trivial relationship between properties and task performance. In the wider context, the framework offers a greater understanding as to what makes a dynamical system compute, helping improve the design of future substrates for RC.