Confabulation dynamics in a reservoir computer: Filling in the gaps with untrained attractors

TL;DR

This paper investigates how confabulation occurs in reservoir computers, revealing that untrained attractors are an intrinsic feature of bounded state space systems and influence learning failures and transitions.

Contribution

It provides a foundational analysis of confabulation in reservoir computers, highlighting the role of untrained attractors in learning failures and system dynamics.

Findings

Untrained attractors are intrinsic to bounded state space systems.

Confabulation influences transitions between reconstructed attractors.

UAs are present in systems beyond reservoir computers.

Abstract

Artificial Intelligence has advanced significantly in recent years thanks to innovations in the design and training of artificial neural networks (ANNs). Despite these advancements, we still understand relatively little about how elementary forms of ANNs learn, fail to learn, and generate false information without the intent to deceive, a phenomenon known as `confabulation'. To provide some foundational insight, in this paper we analyse how confabulation occurs in reservoir computers (RCs): a dynamical system in the form of an ANN. RCs are particularly useful to study as they are known to confabulate in a well-defined way: when RCs are trained to reconstruct the dynamics of a given attractor, they sometimes construct an attractor that they were not trained to construct, a so-called `untrained attractor' (UA). This paper sheds light on the role played by UAs when reconstruction fails and their influence when modelling transitions between reconstructed attractors. Based on our results, we conclude that UAs are an intrinsic feature of learning systems whose state spaces are bounded, and that this means of confabulation may be present in systems beyond RCs.