Multifunctionality in a Reservoir Computer

Abstract

Multifunctionality is a well observed phenomenological feature of biological neural networks and considered to be of fundamental importance to the survival of certain species over time. These multifunctional neural networks are capable of performing more than one task without changing any network connections. In this paper we investigate how this neurological idiosyncrasy can be achieved in an artificial setting with a modern machine learning paradigm known as `Reservoir Computing'. A training technique is designed to enable a Reservoir Computer to perform tasks of a multifunctional nature. We explore the critical effects that changes in certain parameters can have on the Reservoir Computers' ability to express multifunctionality. We also expose the existence of several `untrained attractors'; attractors which dwell within the prediction state space of the Reservoir Computer that were not part of the training. We conduct a bifurcation analysis of these untrained attractors and discuss the implications of our results.

References in corpus (6)

• Recent Advances in Physical Reservoir Computing: A Review
• Using Machine Learning to Replicate Chaotic Attractors and Calculate Lyapunov Exponents from Data
• Attractor Reconstruction by Machine Learning
• Separation of Chaotic Signals by Reservoir Computing
• Using Machine Learning to Assess Short Term Causal Dependence and Infer Network Links
• Bayesian optimization of hyper-parameters in reservoir computing

Cited by in corpus (5)

• Catch-22s of reservoir computing
• How more data can hurt: Instability and regularization in next-generation reservoir computing
• Thermodynamic limit in learning period three
• Denoising and Reconstruction of Nonlinear Dynamics using Truncated Reservoir Computing
• Attractor-merging Crises and Intermittency in Reservoir Computing