Next Generation Reservoir Computing

Overview

Journal Nat Commun

Specialty Biology

Date 2021 Sep 22

PMID 34548491

Citations 45

Authors

Daniel J Gauthier

Erik Bollt

Aaron Griffith

Wendson A S Barbosa

Affiliations

Soon will be listed here.

Abstract

Reservoir computing is a best-in-class machine learning algorithm for processing information generated by dynamical systems using observed time-series data. Importantly, it requires very small training data sets, uses linear optimization, and thus requires minimal computing resources. However, the algorithm uses randomly sampled matrices to define the underlying recurrent neural network and has a multitude of metaparameters that must be optimized. Recent results demonstrate the equivalence of reservoir computing to nonlinear vector autoregression, which requires no random matrices, fewer metaparameters, and provides interpretable results. Here, we demonstrate that nonlinear vector autoregression excels at reservoir computing benchmark tasks and requires even shorter training data sets and training time, heralding the next generation of reservoir computing.

Citing Articles

Toward a physics-guided machine learning approach for predicting chaotic systems dynamics.

Feng L, Liu Y, Shi B, Liu J Front Big Data. 2025; 7:1506443.

PMID: 39897066 PMC: 11782262. DOI: 10.3389/fdata.2024.1506443.

Efficient optimisation of physical reservoir computers using only a delayed input.

Picco E, Jaurigue L, Ludge K, Massar S Commun Eng. 2025; 4(1):3.

PMID: 39827312 PMC: 11742992. DOI: 10.1038/s44172-025-00340-6.

Novel efficient reservoir computing methodologies for regular and irregular time series classification.

Li Z, Andreev A, Hramov A, Blyuss O, Zaikin A Nonlinear Dyn. 2025; 113(5):4045-4062.

PMID: 39822383 PMC: 11732944. DOI: 10.1007/s11071-024-10244-3.

Principled neuromorphic reservoir computing.

Kleyko D, Kymn C, Thomas A, Olshausen B, Sommer F, Frady E Nat Commun. 2025; 16(1):640.

PMID: 39809739 PMC: 11733134. DOI: 10.1038/s41467-025-55832-y.

Ultrafast silicon photonic reservoir computing engine delivering over 200 TOPS.

Wang D, Nie Y, Hu G, Tsang H, Huang C Nat Commun. 2024; 15(1):10841.

PMID: 39738199 PMC: 11686264. DOI: 10.1038/s41467-024-55172-3.

References

Franz M, Scholkopf B . A unifying view of wiener and volterra theory and polynomial kernel regression. Neural Comput. 2006; 18(12):3097-118. DOI: 10.1162/neco.2006.18.12.3097. View

Livi L, Bianchi F, Alippi C . Determination of the Edge of Criticality in Echo State Networks Through Fisher Information Maximization. IEEE Trans Neural Netw Learn Syst. 2017; 29(3):706-717. DOI: 10.1109/TNNLS.2016.2644268. View

Bompas S, Georgeot B, Guery-Odelin D . Accuracy of neural networks for the simulation of chaotic dynamics: Precision of training data vs precision of the algorithm. Chaos. 2020; 30(11):113118. DOI: 10.1063/5.0021264. View

Maass W, Natschlager T, Markram H . Real-time computing without stable states: a new framework for neural computation based on perturbations. Neural Comput. 2002; 14(11):2531-60. DOI: 10.1162/089976602760407955. View

Wikner A, Pathak J, Hunt B, Girvan M, Arcomano T, Szunyogh I . Combining machine learning with knowledge-based modeling for scalable forecasting and subgrid-scale closure of large, complex, spatiotemporal systems. Chaos. 2020; 30(5):053111. DOI: 10.1063/5.0005541. View

Pyle R, Jovanovic N, Subramanian D, Palem K, Patel A . Domain-driven models yield better predictions at lower cost than reservoir computers in Lorenz systems. Philos Trans A Math Phys Eng Sci. 2021; 379(2194):20200246. PMC: 7898131. DOI: 10.1098/rsta.2020.0246. View

Lu Z, Hunt B, Ott E . Attractor reconstruction by machine learning. Chaos. 2018; 28(6):061104. DOI: 10.1063/1.5039508. View

AlMomani A, Sun J, Bollt E . How entropic regression beats the outliers problem in nonlinear system identification. Chaos. 2020; 30(1):013107. DOI: 10.1063/1.5133386. View

Platt J, Wong A, Clark R, Penny S, Abarbanel H . Robust forecasting using predictive generalized synchronization in reservoir computing. Chaos. 2022; 31(12):123118. DOI: 10.1063/5.0066013. View

10.

Bollt E . On explaining the surprising success of reservoir computing forecaster of chaos? The universal machine learning dynamical system with contrast to VAR and DMD. Chaos. 2021; 31(1):013108. DOI: 10.1063/5.0024890. View

11.

Wikner A, Pathak J, Hunt B, Szunyogh I, Girvan M, Ott E . Using data assimilation to train a hybrid forecast system that combines machine-learning and knowledge-based components. Chaos. 2021; 31(5):053114. DOI: 10.1063/5.0048050. View

12.

Brunton S, Proctor J, Kutz J . Discovering governing equations from data by sparse identification of nonlinear dynamical systems. Proc Natl Acad Sci U S A. 2016; 113(15):3932-7. PMC: 4839439. DOI: 10.1073/pnas.1517384113. View

13.

Pathak J, Lu Z, Hunt B, Girvan M, Ott E . Using machine learning to replicate chaotic attractors and calculate Lyapunov exponents from data. Chaos. 2018; 27(12):121102. DOI: 10.1063/1.5010300. View

14.

Griffith A, Pomerance A, Gauthier D . Forecasting chaotic systems with very low connectivity reservoir computers. Chaos. 2020; 29(12):123108. DOI: 10.1063/1.5120710. View

15.

Lai Y . Finding nonlinear system equations and complex network structures from data: A sparse optimization approach. Chaos. 2021; 31(8):082101. DOI: 10.1063/5.0062042. View

16.

Gonon L, Ortega J . Reservoir Computing Universality With Stochastic Inputs. IEEE Trans Neural Netw Learn Syst. 2019; 31(1):100-112. DOI: 10.1109/TNNLS.2019.2899649. View

17.

Pathak J, Hunt B, Girvan M, Lu Z, Ott E . Model-Free Prediction of Large Spatiotemporally Chaotic Systems from Data: A Reservoir Computing Approach. Phys Rev Lett. 2018; 120(2):024102. DOI: 10.1103/PhysRevLett.120.024102. View

18.

Vlachas P, Pathak J, Hunt B, Sapsis T, Girvan M, Ott E . Backpropagation algorithms and Reservoir Computing in Recurrent Neural Networks for the forecasting of complex spatiotemporal dynamics. Neural Netw. 2020; 126:191-217. DOI: 10.1016/j.neunet.2020.02.016. View

19.

Jaeger H, Haas H . Harnessing nonlinearity: predicting chaotic systems and saving energy in wireless communication. Science. 2004; 304(5667):78-80. DOI: 10.1126/science.1091277. View

20.

Thiede L, Parlitz U . Gradient based hyperparameter optimization in Echo State Networks. Neural Netw. 2019; 115:23-29. DOI: 10.1016/j.neunet.2019.02.001. View