Deep time-delay reservoir computing: dynamics and memory capacity

DOI10.1063/5.0017974MaRDI QIDQ5139760zbMATH OpenOpenAlexWikidataFDO

Authors Mirko Goldmann, Felix Köster, Kathy Ludge, S. Yanchuk

Publication date 10 December 2020

Published in Chaos: An Interdisciplinary Journal of Nonlinear Science (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2006.06322

Mathematics Subject Classification ID

Time series analysis of dynamical systems (37M10) Computational methods for ergodic theory (approximation of invariant measures, computation of Lyapunov exponents, entropy, etc.) (37M25)

Abstract: The Deep Time-Delay Reservoir Computing concept utilizes unidirectionally connected systems with time-delays for supervised learning. We present how the dynamical properties of a deep Ikeda-based reservoir are related to its memory capacity (MC) and how that can be used for optimization. In particular, we analyze bifurcations of the corresponding autonomous system and compute conditional Lyapunov exponents, which measure the generalized synchronization between the input and the layer dynamics. We show how the MC is related to the systems distance to bifurcations or magnitude of the conditional Lyapunov exponent. The interplay of different dynamical regimes leads to a adjustable distribution between linear and nonlinear MC. Furthermore, numerical simulations show resonances between clock cycle and delays of the layers in all degrees of the MC. Contrary to MC losses in a single-layer reservoirs, these resonances can boost separate degrees of the MC and can be used, e.g., to design a system with maximum linear MC. Accordingly, we present two configurations that empower either high nonlinear MC or long time linear MC.

Recommendations

• An introduction to delay-coupled reservoir computing
• Memory in linear recurrent neural networks in continuous time
• Revisiting the memory capacity in reservoir computing of directed acyclic network
• Nonlinear memory capacity of parallel time-delay reservoir computers in the processing of multidimensional signals
• Deep reservoir computing

Cites work

• Applied Delay Differential Equations
• Deep reservoir neural networks for trees
• Delay-induced switched states in a slow-fast system
• Enhancing Ikeda time delay system by breaking the symmetry of sine nonlinearity
• Introduction to functional differential equations
• Real-Time Computing Without Stable States: A New Framework for Neural Computation Based on Perturbations
• Spatio-temporal phenomena in complex systems with time delays

Cited in

(14)

• Absolute stability and absolute hyperbolicity in systems with discrete time-delays
• An introduction to delay-coupled reservoir computing
• Revisiting the memory capacity in reservoir computing of directed acyclic network
• Reservoir Computing
• scientific article; zbMATH DE number 7155814 (Why is no real title available?)
• scientific article; zbMATH DE number 7164779 (Why is no real title available?)
• Homeostatic plasticity for single node delay-coupled reservoir computing
• Nonlinear memory capacity of parallel time-delay reservoir computers in the processing of multidimensional signals
• Photonic delay systems as machine learning implementations
• Data-informed reservoir computing for efficient time-series prediction
• Classification of hyperchaotic, chaotic, and regular signals using single nonlinear node delay-based reservoir computers
• Time series forecasting with delay-based reservoir computing: analysis and optimization
• Memory-based reservoir computing for synchronization and its applications
• Disentangling memory and nonlinearity in time-multiplexed optical reservoir computing

This page was built for publication: Deep time-delay reservoir computing: dynamics and memory capacity

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5139760)