Exploring the origins of switching dynamics in a multifunctional reservoir computer

TL;DR

This paper investigates the switching dynamics in multifunctional reservoir computers, revealing how metastability and mode switching arise when reconstructing multiple attractors without external input.

Contribution

It provides a detailed analysis of the origins of switching behavior in reservoir computers when trained on multiple attractors, using the 'seeing double' problem as a case study.

Findings

Switching dynamics are linked to metastability in reservoir computers.

Failure to reconstruct attractors leads to spontaneous mode switching.

The study clarifies the mechanisms behind switching in multifunctional RCs.

Abstract

The concept of multifunctionality has enabled reservoir computers (RCs), a type of dynamical system that is typically realised as an artificial neural network, to reconstruct multiple attractors simultaneously using the same set of trained weights. However there are many additional phenomena that arise when training a RC to reconstruct more than one attractor. Previous studies have found that, in certain cases, if the RC fails to reconstruct a coexistence of attractors then it exhibits a form of metastability whereby, without any external input, the state of the RC switches between different modes of behaviour that resemble properties of the attractors it failed to reconstruct. In this paper we explore the origins of these switching dynamics in a paradigmatic setting via the `seeing double' problem.