Time-shift selection for reservoir computing using a rank-revealing QR algorithm

TL;DR

This paper introduces a novel, task-independent method using a rank-revealing QR algorithm to select optimal time-shifts in reservoir computing, significantly enhancing prediction accuracy without requiring system models.

Contribution

It presents a new technique for time-shift selection in reservoir computing that maximizes reservoir matrix rank, applicable to both analog hardware and traditional networks.

Findings

Improved accuracy over random time-shifts in reservoir computing tasks.

Applicable to both opto-electronic and recurrent neural network reservoirs.

Does not require system modeling or task-specific tuning.

Abstract

Reservoir computing, a recurrent neural network paradigm in which only the output layer is trained, has demonstrated remarkable performance on tasks such as prediction and control of nonlinear systems. Recently, it was demonstrated that adding time-shifts to the signals generated by a reservoir can provide large improvements in performance accuracy. In this work, we present a technique to choose the time-shifts by maximizing the rank of the reservoir matrix using a rank-revealing QR algorithm. This technique, which is not task dependent, does not require a model of the system, and therefore is directly applicable to analog hardware reservoir computers. We demonstrate our time-shift selection technique on two types of reservoir computer: one based on an opto-electronic oscillator and the traditional recurrent network with a $tanh$ activation function. We find that our technique provides improved accuracy over random time-shift selection in essentially all cases.