Transport in reservoir computing

DOI10.1016/J.PHYSD.2023.133744MaRDI QIDQ6098252zbMATH OpenOpenAlexFDO

Authors G. Manjunath, Juan-Pablo Ortega

Publication date 13 June 2023

Published in Physica D (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2209.07946

zbMATH Keywords

driven dynamical systemsFoias operatorrecurrent neural networkreservoir computingstochastic contractiontransport

Mathematics Subject Classification ID

Applications of dynamical systems (37N99) Artificial neural networks and deep learning (68T07)

Abstract: Reservoir computing systems are constructed using a driven dynamical system in which external inputs can alter the evolving states of a system. These paradigms are used in information processing, machine learning, and computation. A fundamental question that needs to be addressed in this framework is the statistical relationship between the input and the system states. This paper provides conditions that guarantee the existence and uniqueness of asymptotically invariant measures for driven systems and shows that their dependence on the input process is continuous when the set of input and output processes are endowed with the Wasserstein distance. The main tool in these developments is the characterization of those invariant measures as fixed points of naturally defined Foias operators that appear in this context and which have been profusely studied in the paper. Those fixed points are obtained by imposing a newly introduced stochastic state contractivity on the driven system that is readily verifiable in examples. Stochastic state contractivity can be satisfied by systems that are not state-contractive, which is a need typically evoked to guarantee the echo state property in reservoir computing. As a result, it may actually be satisfied even if the echo state property is not present.

Recommendations

• scientific article; zbMATH DE number 7306919
• Reservoir computing with an inertial form
• Reservoir computing with computational matter
• On the characteristics and structures of dynamical systems suitable for reservoir computing
• Universal discrete-time reservoir computers with stochastic inputs and linear readouts using non-homogeneous state-affine systems

Cites work

• A fixed-point farrago
• An invitation to statistics in Wasserstein space
• Approximation bounds for random neural networks and reservoir systems
• Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom Inflation
• Decay of correlations
• Discrete-time nonautonomous dynamical systems
• Echo state networks are universal
• Echo state property linked to an input: exploring a fundamental characteristic of recurrent neural networks
• Embedding information onto a dynamical system
• Fading memory and the problem of approximating nonlinear operators with Volterra series
• Fading memory echo state networks are universal
• Generalized autoregressive conditional heteroscedasticity
• scientific article; zbMATH DE number 3134565 (Why is no real title available?)
• scientific article; zbMATH DE number 976356 (Why is no real title available?)
• scientific article; zbMATH DE number 7306919 (Why is no real title available?)
• scientific article; zbMATH DE number 7164779 (Why is no real title available?)
• scientific article; zbMATH DE number 5169461 (Why is no real title available?)
• Liquid state machines: motivation, theory, and applications
• Memory loss for nonequilibrium open dynamical systems
• Memory loss for time-dependent piecewise expanding systems in higher dimension
• Non-stationary compositions of Anosov diffeomorphisms
• Optimal Transport
• Polynomial loss of memory for maps of the interval with a neutral fixed point
• Real-Time Computing Without Stable States: A New Framework for Neural Computation Based on Perturbations
• Reservoir computing approaches to recurrent neural network training
• Stability and memory-loss go hand-in-hand: three results in dynamics and computation
• Statistical properties of dynamical systems with some hyperbolicity
• Strict stationarity of generalized autoregressive processes
• The development of chaotic advection
• The stochastic equation Y_n+1=A_nY_n + B_n with stationary coefficients
• Universal discrete-time reservoir computers with stochastic inputs and linear readouts using non-homogeneous state-affine systems

Cited in

(2)

• Non-instantaneous information transfer in physical reservoir computing
• Forecasting causal dynamics with universal reservoirs

This page was built for publication: Transport in reservoir computing

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6098252)