Harnessing Nonlinearity: Predicting Chaotic Systems and Saving Energy in Wireless Communication
The DOI matches its Crossref record; 18 of the 18 references it cites were found in a registry or an index.
Abstract
We present a method for learning nonlinear systems, echo state networks (ESNs). ESNs employ artificial recurrent neural networks in a way that has recently been proposed independently as a learning mechanism in biological brains. The learning method is computationally efficient and easy to use. On a benchmark task of predicting a chaotic time series, accuracy is improved by a factor of 2400 over previous techniques. The potential for engineering applications is illustrated by equalizing a communication channel, where the signal error rate is improved by two orders of magnitude.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Although work in artificial intelligence and machine-learning research has shown that similar or even more complex outcomes can be achieved with a variety of techniques, including deep neural networks with trained weights (Gibney, 2015), reservoir networks with a trained output mapping (Maass et al., 2002;Jaeger & Haas, 2004), networks that are evolved using a genetic algorithm (Iizuka & Di Paolo,”
From A potential mechanism for Gibsonian resonance: behavioral entrainment emerges from local homeostasis in an unsupervised reservoir network (2024) · cites this paper as Jaeger & Haas, 2004
In the citing paper
“Therefore, a sequence production solely by a recurrent neural network can be unstable against small fluctuations in activities and unwanted noises (Jaeger and Haas, 2004).”
From The Cerebro-Cerebellum as a Locus of Forward Model: A Review (2020) · cites this paper as Jaeger and Haas, 2004
Citations
Cites
Backpropagation through time: what it does and how to do it
Werbos, P. J. (1990) · Proceedings of the IEEE · doi:10.1109/5.58337 · Cited by 4,879
A Learning Algorithm for Continually Running Fully Recurrent Neural Networks
Williams, R. J., & Zipser, D. (1989) · Neural Computation · doi:10.1162/neco.1989.1.2.270 · Cited by 4,369
Oscillation and Chaos in Physiological Control Systems
Mackey, M. C., & Glass, L. (1977) · Science · doi:10.1126/science.267326 · Cited by 4,135
Maass, W., Natschläger, T., & Markram, H. (2002) · Neural Computation · doi:10.1162/089976602760407955 · Cited by 4,025
Real-time prediction of hand trajectory by ensembles of cortical neurons in primates
Wessberg, J., Stambaugh, C. R., Kralik, J. D., Beck, P. D., Laubach, M., Chapin, J. K., Kim, J., Biggs, S. J., Srinivasan, M. A., & Nicolelis, M. A. L. (2000) · Nature · doi:10.1038/35042582 · Cited by 1,487
'Neural-gas' network for vector quantization and its application to time-series prediction
Martinetz, T. M., Berkovich, S. G., & Schulten, K. J. (1993) · IEEE Transactions on Neural Networks · doi:10.1109/72.238311 · Cited by 1,431
Approximation of dynamical systems by continuous time recurrent neural networks
Funahashi, K.-I., & Nakamura, Y. (1993) · Neural Networks · doi:10.1016/s0893-6080(05)80125-x · Cited by 980
A new evolutionary system for evolving artificial neural networks
Yao, X., & Liu, Y. (1997) · IEEE Transactions on Neural Networks · doi:10.1109/72.572107 · Cited by 874
Applying LSTM to Time Series Predictable through Time-Window Approaches
Gers, F. A., Eck, D., & Schmidhuber, J. (2001) · Lecture Notes in Computer Science · doi:10.1007/3-540-44668-0_93 · Cited by 334
Computational neuroscience: a comprehensive approach
(2004) · Choice Reviews Online · doi:10.5860/choice.41-5284 · Cited by 279
Research works printed but not linked: 10
- K. Doya, in The Handbook of Brain Theory and Neural Networks, M. A. Arbib, Ed. (MIT Press, Cambridge, MA, 1995), pp. 796–800.
- J. Vesanto in Proc. WSOM'97 (1997); www.cis.hut.fi/projects/monitor/publications/papers/wsom97.ps.
- L. Chudy, I. Farkas, Neural Network World8, 481 (1998).
- H. Bersini M. Birattari G. Bontempi in Proc. IEEE World Congr. on Computational Intelligence (IJCNN'98) (1997) pp. 2102–2106; ftp://iridia.ulb.ac.be/pub/lazy/papers/IridiaTr1997-13_2.ps.gz.
- V. J. Mathews, J. Lee, in Advanced Signal Processing: Algorithms, Architectures, and Implementations V (Proc. SPIE Vol. 2296), (SPIE, San Diego, CA, 1994), pp. 317–327.
- J. Hertzberg H. Jaeger F. Schönherr in Proc. 15th Europ. Conf. on Art. Int. (ECAI 02) F. van Harmelen Ed. (IOS Press Amsterdam 2002) pp. 708–712; www.ais.fhg.de/∼schoenhe/papers/ECAI02.pdf.
- H. Jaeger, in Advances in Neural Information Processing Systems 15, S. Becker, S. Thrun, K. Obermayer, Eds. (MIT Press, Cambridge, MA, 2003) pp. 593–600.
- G. E. Hinton, in Parallel Models of Associative Memory, G. E. Hinton, J. A. Anderson, Eds. (Erlbaum, Hillsdale, NJ, 1981), pp. 161–187.
- M. Kawato, in The Handbook of Brain Theory and Neural Networks, M. Arbib, Ed. (MIT Press, Cambridge, MA, 1995), pp. 172–178.
- G. B. Stanley, Neurocomputing38–40, 1703 (2001).
Cited by
Items 701–800 of 1,009
Reservoir computing benchmarks: a tutorial review and critique
Wringe, C., Trefzer, M. A., & Stepney, S. (2025) · International Journal of Parallel Emergent and Distributed Systems · doi:10.1080/17445760.2025.2472211 · Cited by 77
“For classification into a larger number of categories, symbol error rate (the fraction of incorrect symbolsobtained)[24]maybeused.”
InfluentialInferring Depression and Affect from Application Dependent Meta Knowledge
Kächele, M., Schels, M., & Schwenker, F. (2014) · Proceedings of the 4th International Workshop on Audio/Visual Emotion Challenge · doi:10.1145/2661806.2661813 · Cited by 44
“Popular choices include recurrent neural networks that are not a.ected by the vanishing gradient problem [15] which can be trained over longer time periods such as Long Short-Term Memory Networks (LSTM) [16] and Echo State Networks (ESN) [18].”
BackgroundEvolving Hardware with Self-reconfigurable connectivity in Xilinx FPGAs
Upegui, A., & Sánchez, E. G. (2006) · NASA/ESA Conference on Adaptive Hardware and Systems (AHS) · doi:10.1109/ahs.2006.38 · Cited by 44
“Jaeger and Haas [9] have shown the suitability of echo state networks –randomly connected recurrent ANNs– to predict chaotic time series, improving accuracy by 2400 over previous techniques.”
MethodInfluentialEncoding Time in Feedforward Trajectories of a Recurrent Neural Network Model
Hardy, N. F., & Buonomano, D. V. (2017) · Neural Computation · doi:10.1162/neco_a_01041 · Cited by 46
“The network dynamics were governed by the standard firing rate equations (Abeles, 1982; Sompolinsky et al., 1988; Jaeger & Haas, 2004): τ dxi dt = −xi + NRec∑ j=1 WReci j r j + NIn∑ k=1 WInik yk + INoisei , (4.1) where xi represents the state of unit i.”
MethodInfluentialData-driven prediction and analysis of chaotic origami dynamics
Yasuda, H., Yamaguchi, K., Miyazawa, Y., Wiebe, R., Raney, J. R., & Yang, J. (2020) · Communications Physics · doi:10.1038/s42005-020-00431-0 · Cited by 47
Flapping‐Wing Dynamics as a Natural Detector of Wind Direction
Tanaka, K., Yang, S.-H., Tokudome, Y., Minami, Y., Lu, Y., Arie, T., Akita, S., Takei, K., & Nakajima, K. (2020) · Advanced Intelligent Systems · doi:10.1002/aisy.202000174 · Cited by 46
Fujii, K., & Nakajima, K. (2021) · Natural computing series · doi:10.1007/978-981-13-1687-6_18 · Cited by 106
Robust forecasting using predictive generalized synchronization in reservoir computing
Platt, J. A., Wong, A., Clark, R., Penny, S. G., & Abarbanel, H. D. I. (2021) · Chaos An Interdisciplinary Journal of Nonlinear Science · doi:10.1063/5.0066013 · Cited by 62
“A class of RNN architectures with demonstrated capability for forecasting dynamical systems is reservoir computing (RC) [3, 4, 5, 6, 7, 8, 9, 10, 11, 12].”
BackgroundComputational role of structure in neural activity and connectivity
Ostojic, S., & Fusi, S. (2024) · Trends in Cognitive Sciences · doi:10.1016/j.tics.2024.03.003 · Cited by 60
Reservoir-computing based associative memory and itinerancy for complex dynamical attractors
Kong, L.-W., Brewer, G. A., & Lai, Y. (2024) · Nature Communications · doi:10.1038/s41467-024-49190-4 · Cited by 60
“Reservoir computing 11 – 13 , a class of recurrent neural networks (RNNs) that has been extensively investigated for model-free prediction of chaotic systems in recent years 14 – 38 , stands out as a suitable choice.”
Sen, D., Ravichandran, H., Das, M. B., Venkatram, P., Choo, S., Varshney, S., Zhang, Z., Sun, Y., Shah, J. M., Radhakrishnan, S. S., et al. (2024) · Nature Communications · doi:10.1038/s41467-024-54231-z · Cited by 48
“…switching in the STO-gated MoS 2 FET to construct a physical reservoir for accelerating RC. Note that, RC, derived from the framework of recurrent neural networks (RNNs), has emerged as a powerful computational paradigm for processing time-series data and performing pattern recognition 30,31 .”
Reinhart, F., & Steil, J. J. (2009) · IEEE-RAS International Conference on Humanoid Robots (Humanoids) · doi:10.1109/ichr.2009.5379558 · Cited by 42
“However, for autonomous generation of movements, usually outputs of the reservoir network are recursively connected to the input, which is also the standard approach for generating complex dynamical behavior in recursive time series prediction [10], [17], [18].”
BackgroundCunillera, A., Soriano, M. C., & Fischer, I. (2019) · Chaos An Interdisciplinary Journal of Nonlinear Science · doi:10.1063/1.5120822 · Cited by 43
Using multidimensional speckle dynamics for high-speed, large-scale, parallel photonic computing
Sunada, S., Kanno, K., & Uchida, A. (2020) · Optics Express · doi:10.1364/oe.399495 · Cited by 52
Discrete-Time Signatures and Randomness in Reservoir Computing
Cuchiero, C., Gonon, L., Grigoryeva, L., Ortega, J., & Teichmann, J. (2021) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2021.3076777 · Cited by 50
“RC can be found in the literature under other denominations, such as liquid state machines [1]–[5], and is represented by various learning paradigms, with echo state networks (ESNs) [6]–[8] being a particularly important example.”
BackgroundChaos-Based Secure Communications in Biomedical Information Application
Liao, T., Chen, H.-C., Peng, C., & Hou, Y. (2021) · Electronics · doi:10.3390/electronics10030359 · Cited by 42
On the Use of Reservoir Computing in Popularity Prediction
Wu, T., Timmers, M., Vleeschauwer, D. D., & Leekwijck, W. V. (2010) · International Conference on Evolving Internet (INTERNET) · doi:10.1109/internet.2010.13 · Cited by 44
“However, on certain days the cumulative views could not be retrieved for some videos, due to unavailability of the video itself or its statistics, leading to incomplete traces with significant gaps.”
BackgroundTemporal-Kernel Recurrent Neural Networks
Sutskever, I., & Hinton, G. E. (2009) · Neural Networks · doi:10.1016/j.neunet.2009.10.009 · Cited by 44
“ESNs work extremely well for predicting chaotic systems for a large number of timesteps, and their linear version has been rigorously analyzed (White et al., 2004).”
MethodNa, X., Ren, W., Liu, M., & Han, M. (2022) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2022.3157830 · Cited by 42
“Meanwhile, ESN [11], ϕ -ESN [29], R 2 SP [30], DeepESN [33], ADRC [34], and DeePr-ESN [35] are selected as the comparison models.”
MethodTemporal information processing induced by quantum noise
Kubota, T., Suzuki, Y., Kobayashi, S., Tran, Q. H., Yamamoto, N., & Nakajima, K. (2023) · Physical Review Research · doi:10.1103/physrevresearch.5.023057 · Cited by 69
Reinforcement Twinning: From digital twins to model-based reinforcement learning
Schena, L., Marques, P. A. D. M., Poletti, R., Ahizi, S., den Berghe, J. V., & Mendez, M. A. (2024) · Journal of Computational Science · doi:10.1016/j.jocs.2024.102421 · Cited by 54
On the Quantification of Dynamics in Reservoir Computing
Verstraeten, D., & Schrauwen, B. (2009) · Lecture notes in computer science · doi:10.1007/978-3-642-04274-4_101 · Cited by 50
Mote-Based Online Anomaly Detection Using Echo State Networks
Chang, M., Terzis, A., & Bonnet, P. (2009) · Lecture notes in computer science · doi:10.1007/978-3-642-02085-8_6 · Cited by 41
“showed that an ESN with a 1000 neuron reservoir can predict over 1000 steps of this time series, (depending on the accuracy) when trained and initialized to a certain point in the MG time series [26].”
BackgroundInformation dynamics based self-adaptive reservoir for delay temporal memory tasks
Dasgupta, S., Wörgötter, F., & Manoonpong, P. (2013) · Evolving Systems · doi:10.1007/s12530-013-9080-y · Cited by 45
“Although both spiking and analog neurons have been explored previously, here we focus on the Echo-state network (ESN) type (Jaeger and Haas 2004) using sigmoid leaky integrator neurons.”
BackgroundShi, G., Liu, D., & Wei, Q. (2017) · IET Control Theory and Applications · doi:10.1049/iet-cta.2016.0653 · Cited by 41
A learning system for automatic Berg Balance Scale score estimation
Bacciu, D., Chessa, S., Gallicchio, C., Micheli, A., Pedrelli, L., Ferro, E., Fortunati, L., Rosa, D. L., Palumbo, F., Vozzi, F., et al. (2017) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2017.08.018 · Cited by 46
“In particular, as a result of an extensive and comprehensive comparison among different learning models, we take into consideration the Reservoir Computing (RC) paradigm (Lukoševičius and Jaeger, 2009; Verstraeten et al., 2007) and specifically the Echo State Network (ESN) model (Jaeger and Haas, 2004; Jaeger, 2001),…”
InfluentialLeast squares support vector machine with self-organizing multiple kernel learning and sparsity
Liu, C., Tang, L., & Liu, J. (2018) · Neurocomputing · doi:10.1016/j.neucom.2018.11.067 · Cited by 43
“3) Comparison with state-of-the-art methods: Experiments were also conducted with other state-of-the-art methods, SVR [45], RVM [46], KELM [47] and ESN [48], to illustrate the effectiveness of the SDE-MKLSSVM.”
MethodBayesian Filtering with Multiple Internal Models: Toward a Theory of Social Intelligence
Isomura, T., Parr, T., & Friston, K. (2019) · Neural Computation · doi:10.1162/neco_a_01239 · Cited by 38
Reservoir Computing with Neuromemristive Nanowire Networks
Fu, K., Zhu, R., Loeffler, A., Hochstetter, J., Álvarez, A. D., Stieg, A. Z., Gimzewski, J. K., Nakayama, T., & Kuncic, Z. (2020) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn48605.2020.9207727 · Cited by 54
Optimizing memory in reservoir computers
Carroll, T. L. (2022) · Chaos An Interdisciplinary Journal of Nonlinear Science · doi:10.1063/5.0078151 · Cited by 60
Hilbert space as a computational resource in reservoir computing
Kalfus, W. D., Ribeill, G. J., Rowlands, G. E., Krovi, H., Ohki, T., & Govia, L. C. G. (2022) · Physical Review Research · doi:10.1103/physrevresearch.4.033007 · Cited by 55
Nonlocal Detection of Interlayer Three-Magnon Coupling
Sheng, L., Elyasi, M., Chen, J., He, W., Wang, Y., Wang, H., Feng, H., Zhang, Y., Medlej, I., Liu, S., et al. (2023) · Physical Review Letters · doi:10.1103/physrevlett.130.046701 · Cited by 41
“rich physics [33] and is relevant for technological applications [34], such as mechanical force sensors [35].”
BackgroundBayesian optimization of hyper-parameters in reservoir computingPreprint
Yperman, J., & Becker, T. (2016) · arXiv (Cornell University) · doi:10.48550/arxiv.1611.05193 · Cited by 53
“We also used an ESN on this data set, and compare our results with [26].”
BackgroundMethodResultInfluentialMeta-Learning through Hebbian Plasticity in Random Networks
Najarro, E., & Risi, S. (2020) · IT University Of Copenhagen (IT University of Copenhagen) · doi:10.48550/arxiv.2007.02686 · Cited by 97
“While recent work in this area has focused on determining both the weights of the network and the plasticity parameters, we are particularly intrigued by the interesting properties of randomly-initialised networks in both machine learning [16–18] and neuroscience [19].”
BackgroundMultiplex visibility graphs to investigate recurrent neural network dynamics
Bianchi, F. M., Livi, L., Alippi, C., & Jenssen, R. (2017) · Scientific Reports · doi:10.1038/srep44037 · Cited by 41
Sensory Adaptation in Biomolecular Memristors Improves Reservoir Computing Performance
Maraj, J. J., Haughn, K. P. T., Inman, D. J., & Sarles, S. A. (2023) · Advanced Intelligent Systems · doi:10.1002/aisy.202300049 · Cited by 42
MINN: Learning the Dynamics of Differential-Algebraic Equations and Application to Battery Modeling
Huang, Y., Zou, C., Li, Y., & Wik, T. (2024) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2024.3456475 · Cited by 56
“To blend neural networks with governing physical laws directly, the physics-informed neural networks (PINNs) have been introduced to approximate solutions to PDEs by incorporating physical constraints into the loss function [30].”
Feedback-Driven Quantum Reservoir Computing for Time-Series Analysis
Kobayashi, K., Fujii, K., & Yamamoto, N. (2024) · PRX Quantum · doi:10.1103/prxquantum.5.040325 · Cited by 99
“For performance evaluation of the QRC in predicting classical dynamics, we utilize the Mackey-Glass (MG) delay differential equation [13, 16, 53, 54]: With the parameters α = 0 .”
Cerebellar-driven cortical dynamics can enable task acquisition, switching and consolidation
Pemberton, J., Chadderton, P. T., & Costa, R. P. (2024) · Nature Communications · doi:10.1038/s41467-024-55315-6 · Cited by 47
“1b) 14,16 and with feedback provided by a cortico-cerebellar loop (Fig.”
InfluentialOptical next generation reservoir computing
Wang, H., Hu, J., Baek, Y., Tsuchiyama, K., Joly, M., Liu, Q., & Gigan, S. (2025) · Light Science & Applications · doi:10.1038/s41377-025-01927-6 · Cited by 62
“This potential has been clearly evidenced by several previous experimental studies, where the benefits of optical computing emerge as N approaches the order of 10 4 [4, 7, 21].”
A simple feedforward convolutional conceptor neural network for classification
Qian, G., & Zhang, L. (2017) · Applied Soft Computing · doi:10.1016/j.asoc.2017.08.016 · Cited by 42
Qiao, J., Wang, L., & Yang, C. (2018) · Neural Computing and Applications · doi:10.1007/s00521-018-3420-6 · Cited by 42
“An original ESN (OESN) [11] without feedback connections has three layers, the input layer, the reservoir layer and the output layer, as shown in Fig.”
BackgroundMethodInfluentialAlizamir, M., Kim, S., Kişi, Ö., & Zounemat‐Kermani, M. (2020) · Hydrological Sciences Journal · doi:10.1080/02626667.2020.1735639 · Cited by 42
“The ESN is an e ffi cient implementation of a recurrent neural network (RNN) within the reservoir computing (RC) framework (Jaeger and Haas 2004).”
BackgroundLiu, C., Zhang, H., Luo, Y., & Su, H. (2020) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2020.2984952 · Cited by 40
Li, Q., Wu, Z., Ling, R., Feng, L., & Liu, K. (2020) · Applied Soft Computing · doi:10.1016/j.asoc.2020.106481 · Cited by 42
Predicting phase and sensing phase coherence in chaotic systems with machine learning
Zhang, C., Jiang, J., Xian, S., & Lai, Y. (2020) · Chaos An Interdisciplinary Journal of Nonlinear Science · doi:10.1063/5.0006304 · Cited by 42
“1,3 21 In general, there are three main components in a reservoir computing machine:”
BackgroundPreliminary Research of Chaotic Characteristics and Prediction of Short-Term Wind Speed Time Series
Tian, Z. (2020) · International Journal of Bifurcation and Chaos · doi:10.1142/s021812742050176x · Cited by 40
Reservoir Computing Approach to Quantum State Measurement
Angelatos, G., Khan, S. A., & Türeci, H. E. (2021) · Physical Review X · doi:10.1103/physrevx.11.041062 · Cited by 67
“Reservoir computing is a machine learning framework for the processing of time-dependant data [17–20].”
MethodSelf-organizing deep belief modular echo state network for time series prediction
Zhang, H., Hu, B., Wang, X., Xu, J., Wang, L., Sun, Q., & Wang, Z. (2021) · Knowledge-Based Systems · doi:10.1016/j.knosys.2021.107007 · Cited by 40
A New Hypothesis for Sleep: Tuning for Criticality
Pearlmutter, B. A., & Houghton, C. J. (2009) · Neural Computation · doi:10.1162/neco.2009.05-08-787 · Cited by 38
“As we have discussed, criticality has engineering advantages; furthermore, some learning algorithms for temporal sequences, such as liquid state machines (Maass, Natschlager, & Markram, 2002 ) and echo state networks (Jaeger & Haas, 2004 ), perform well only when their recurrent network modules are near the boundary o…”
BackgroundRegularization and nonlinearities for neural language models: when are they needed?Preprint
Pachitariu, M., & Sahani, M. (2013) · arXiv (Cornell University) · doi:10.48550/arxiv.1301.5650 · Cited by 52
“The analysis of [10] also suggests momentum as an important technique for training RNNs and deep neural networks, although the authors suggests it should be used in conjunction with a special initialization of the RNN as an Echo State Network [11] but we did not investigate such an initialization.”
MethodLearning Hamiltonian dynamics with reservoir computing
Zhang, H., Fan, H., Wang, L., & Wang, X. (2021) · Physical review. E · doi:10.1103/physreve.104.024205 · Cited by 52
“For instance, it is demonstrated that a well trained RC can accurately predict the state evolution of chaotic systems for about a half-dozen Lyapunov times, which is much longer than the prediction horizon of conventional techniques [6].”
BackgroundAsynchronous photonic time-delay reservoir computing
Tang, J.-Y., Lin, B.-D., Shen, Y.-W., Li, R., Yu, J., He, X., & Wang, C. (2022) · Optics Express · doi:10.1364/oe.478728 · Cited by 43
Wang, J., Wu, C., & Niu, T. (2019) · Sustainability · doi:10.3390/su11020526 · Cited by 42
“Echo state network (ESN), which has shown successful applications to time series forecasting like load forecasting, stock forecasting, energy consumption forecasting, and so on, can deal with the difficulties mentioned above but is still rarely applied in wind speed time series forecasting [28–32].”
BackgroundYue, D., Wu, Z.-M., Hou, Y., Cui, B., Jin, Y.-H., Dai, M., & Xia, G.-Q. (2019) · Optics Express · doi:10.1364/oe.27.019931 · Cited by 43
Xu, S., Li, J., Liu, K., & Wu, L. (2019) · IEEE Access · doi:10.1109/access.2019.2936516 · Cited by 43
“These large-scale randomly connected recurrent networks are used to replace the middle layer of classical neural networks in order to simplify the network training process [21].”
MethodThe reservoir’s perspective on generalized synchronization
Lymburn, T., Walker, D. M., SMALL, M. P., & Jüngling, T. (2019) · Chaos An Interdisciplinary Journal of Nonlinear Science · doi:10.1063/1.5120733 · Cited by 50
Parallel Time-Delay Reservoir Computing With Quantum Dot Lasers
Tang, J.-Y., Lin, B.-D., Yu, J., He, X., & Wang, C. (2022) · IEEE Journal of Quantum Electronics · doi:10.1109/jqe.2022.3146561 · Cited by 41
Artificial Intelligence in Classical and Quantum Photonics
Vernuccio, F., Bresci, A., Cimini, V., Giuseppi, A., Cerullo, G., Polli, D., & Valensise, C. M. (2022) · Laser & Photonics Review · doi:10.1002/lpor.202100399 · Cited by 46
Self-organization of an inhomogeneous memristive hardware for sequence learning
Payvand, M., Moro, F., Nomura, K., Dalgaty, T., Vianello, E., Nishi, Y., & Indiveri, G. (2022) · Nature Communications · doi:10.1038/s41467-022-33476-6 · Cited by 49
“As the sequence length increases, the networkneedstorememberincreasinglylongersequenceswhichtests its fading memory 36 .”
Energy-Efficient Neuromorphic Classifiers
Marti, D. N., Rigotti, M., Seok, M., & Fusi, S. (2016) · Neural Computation · doi:10.1162/neco_a_00882 · Cited by 41
“The combination of a random mixing of the inputs together with a non-linear input-output transformation efficiently expands the dimensionality of the resulting signal (see e.g. [7, 8, 9]), thereby increasing the chances that downstream neurons can 1 – 12 a r X i v : 1507 .”
BackgroundTowards Comprehensive Foundations of Computational Intelligence
Duch, W. (2007) · Studies in computational intelligence · doi:10.1007/978-3-540-71984-7_11 · Cited by 43
“These approaches are equivalent to the Liquid State Machines [84] and are being applied to time series prediction, dynamical system identification and speech recognition with great success [84].”
MethodAn associative memory readout for ESNs with applications to dynamical pattern recognition
Ozturk, M. C., & Príncipe, J. C. (2007) · Neural Networks · doi:10.1016/j.neunet.2007.04.012 · Cited by 39
Palacz, A., John, M. S., Brewin, R. J. W., Hirata, T., & Gregg, W. W. (2013) · Biogeosciences · doi:10.5194/bg-10-7553-2013 · Cited by 42
“ANNs are already heavily relied upon in system control and management applications in other disciplines such as electrical engineering, medical science or business and economics (Anandarajan et al., 2001; Jaeger and Haas, 2004; Khan et al., 2001).”
BackgroundNeuromorphic Photonics, Principles of
Shastri, B. J., Tait, A. N., de Lima, T. F., Nahmias, M. A., Peng, H.-T., & Prucnal, P. R. (2018) · Springer eBooks · doi:10.1007/978-3-642-27737-5_702-1 · Cited by 52
Highly Accurate Gear Fault Diagnosis Based on Support Vector Machine
Abdul, Z. K., & Al‐Talabani, A. K. (2022) · Journal of Vibration Engineering & Technologies · doi:10.1007/s42417-022-00768-6 · Cited by 38
“Another popular progress of the RNN model is Echo State Network (ESN), which was invented by Jaeger and Haas [45].”
MethodManifold clustering-based prediction for dynamic multiobjective optimization
Li, Y., Qi, W., Qin, A. K., Yang, S., Gong, D., Qu, B., & Liang, J. (2023) · Swarm and Evolutionary Computation · doi:10.1016/j.swevo.2023.101254 · Cited by 40
In-sensor reservoir computing based on optoelectronic synaptic devices
Chen, Z., Xiao, Y., Huang, W.-Y., Jiang, Y., Liu, Q., & Tang, X. (2023) · Applied Physics Letters · doi:10.1063/5.0160599 · Cited by 39
Deep Learning for Dynamic Graphs: Models and Benchmarks
Gravina, A., & Bacciu, D. (2024) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2024.3379735 · Cited by 72
“Specifically, the authors propose to encode the temporal dynamic by means of an ESNs [54], [55], a randomized RNNs, 3 to efficiently compute node embedding and improve the scalability of DGNs for D-TDG.”
BackgroundDNA Reservoir Computing: A Novel Molecular Computing Approach
Goudarzi, A., Lakin, M. R., & Stefanović, D. (2013) · Lecture notes in computer science · doi:10.1007/978-3-319-01928-4_6 · Cited by 39
“A reservoir computer is a device that uses transient dynamics of a system in a critical regime—a regime in which perturbations to the system’s trajectory in its phase space neither spread nor die out—to transform an input signal into a desired output [1].”
BackgroundUtilizing online learning based on echo-state networks for the control of a hydraulic excavator
Park, J., Cho, D., Kim, S.-H., Kim, Y. B., Kim, P. Y., & Kim, H. J. (2014) · Mechatronics · doi:10.1016/j.mechatronics.2014.10.004 · Cited by 41
Initialization and self‐organized optimization of recurrent neural network connectivity
Boedecker, J., Obst, O., Mayer, N. M., & Asada, M. (2009) · HFSP journal · doi:10.2976/1.3240502 · Cited by 44
“ESN provide a specific architecture and a training procedure that aims to solve the problem of slow convergence (Jaeger, 2001a; Jaeger and Haas, 2004).”
BackgroundSemi-supervised Echo State Networks for Audio Classification
Scardapane, S., & Uncini, A. (2016) · Cognitive Computation · doi:10.1007/s12559-016-9439-z · Cited by 43
Deep reservoir computing: A critical analysis
Gallicchio, C., & Micheli, A. (2016) · CINECA IRIS Institutial research information system (University of Pisa) · Cited by 38
Difference between memory and prediction in linear recurrent networks
Marzen, S. (2017) · Physical review. E · doi:10.1103/physreve.96.032308 · Cited by 38
“Such is the case, it seems, in the field of echo state networks (ESNs) [1,2].”
Cao, F., Yang, Z., Ren, J., Ling, W.-K., Zhao, H., & Marshall, S. (2017) · Remote Sensing · doi:10.3390/rs9121255 · Cited by 39
“In fact, besides ELM, some other models, such as liquid state machines [35], echo state networks [36-37], also have adopted the random weight selection technique with great success [38].”
MethodMolecules, semiconductors, light and information: Towards future sensing and computing paradigms
Pilarczyk, K., Wlaźlak, E., Przyczyna, D., Blachecki, A., Podborska, A., Athanasiou, V., Konkoli, Z., & Szacilowski, K. T. (2018) · Coordination Chemistry Reviews · doi:10.1016/j.ccr.2018.03.018 · Cited by 40
Echo State Network for the Remaining Useful Life Prediction of a Turbofan Engine
Rigamonti, M., Baraldi, P., Zio, E., Roychoudhury, I., Goebel, K., & Poll, S. D. (2016) · PHM Society European Conference · doi:10.36001/phme.2016.v3i1.1623 · Cited by 38
“Although ESNs have been extensively investigated and used for the prediction of chaotic time series (Jaeger & Haas, 2004; Shi & Han, 2007; Li, Han, & Wang, 2012), they have so far seen limited use for RUL prediction of industrial systems (Peng, Wang., Wang, Liu, & Peng, 2012), (Morando, Jemei,…”
BackgroundNguimdo, R. M., & Erneux, T. (2018) · Optics Letters · doi:10.1364/ol.44.000049 · Cited by 38
Singular Value Decomposition (SVD)
Brunton, S. L., & Kutz, J. N. (2019) · Cambridge University Press eBooks · doi:10.1017/9781108380690.002 · Cited by 74
Na, X., Han, M., Ren, W., & Zhong, K. (2020) · IEEE Transactions on Cybernetics · doi:10.1109/tcyb.2020.2977375 · Cited by 38
“Currently, the echo state network (ESN), a kind of recurrent neural network with a large number of randomly and sparsely generated neurons, has been widely used in time-series prediction [8], [9].”
BackgroundTailoring Echo State Networks for Optimal Learning
Aceituno, P. V., Yan, G. F., & Liu, Y. (2020) · iScience · doi:10.1016/j.isci.2020.101440 · Cited by 38
“To further demonstrate the validity of CjljD as a proxy measure for the ESN performance, we consider the following tasks (Figures 3A–3C): (1) forecasting the chaotic Mackey-Glass time series (Mackey and Glass, 1977), which is a benchmark task to evaluate the performance of ESN (Jaeger and Haas, 2004; Luko sevicius and…”
BackgroundMethodInfluentialGuo, X., Xiang, S., Qu, Y., Han, Y. N., Wen, A. J., & Hao, Y. (2020) · Journal of Lightwave Technology · doi:10.1109/jlt.2020.3023451 · Cited by 38
Twin vortex computer in fluid flow
Goto, K., Nakajima, K., & Notsu, H. (2021) · New Journal of Physics · doi:10.1088/1367-2630/ac024d · Cited by 47
“Furthermore, we compare the performance of our system with a simple linear regression (LR) model, y k = w r k + w , which bypassed the reservoir part and directly connected the input to the readout, clarifying the contribution of the reservoir, and with an echo-state network (ESN) [10], which is the standard recurrent…”
BackgroundMethodInfluentialPhotonic neuromorphic technologies in optical communications
Argyris, A. (2022) · Nanophotonics · doi:10.1515/nanoph-2021-0578 · Cited by 52
“A rather speci fi c task, that was studied fi rst in [121] for wireless transmission, resulted as a popular benchmark task for the echo-state network community [83].”
Photonic reservoir computer based on frequency multiplexing
Butschek, L., Akrout, A., Dimitriadou, E., Lupo, A., Haelterman, M., & Massar, S. (2022) · Optics Letters · doi:10.1364/ol.451087 · Cited by 50
“We report here results on two benchmark tasks: nonlinear channel equalization, following the model reported in [8], and a time-series prediction of the recorded output of a chaotically operating far-infrared laser (Santa-Fe) [33] (results on other tasks can be found in [26]).”
BackgroundPrediction of the Italian electricity price for smart grid applications
Crisostomi, E., Gallicchio, C., Micheli, A., Raugi, M., & Tucci, M. (2015) · Neurocomputing · doi:10.1016/j.neucom.2015.02.089 · Cited by 40
“The linear readout is the only trained part, typically by Moore-Penrose pseudo-inversion or ridge regression [12, 11].”
BackgroundMethodInfluentialQuevedo, J., Chen, H., Escofet, M. À. C., Peter Tiňo, B., Puig, V., García, D., Sarrate, R., & Yao, X. (2014) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2014.01.008 · Cited by 39
“Jaeger and Haas [32] suggested that reservoirs should ideally have small clustering degree (sparse reservoirs) to spread the dynamic information in the reservoir and a small average path length can be beneficial for a variety of dynamical time scales.”
Time Warping Invariant Echo State Networks
Lukoševičius, M., Popovici, D., Jaeger, H., & Siewert, U. (2006) · Cited by 38
Self-citationComparison between DeepESNs and gated RNNs on multivariate time-series predictionPreprint
Gallicchio, C., Micheli, A., & Pedrelli, L. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1812.11527 · Cited by 43
“Training of the readout was performed through ridge regression [9, 3] with regularization coefficient λr in {10, 10, 10, 10}.”
The reservoir learning power across quantum many-body localization transition
Xia, W., Zou, J., Qiu, X., & Li, X. (2022) · Frontiers of Physics · doi:10.1007/s11467-022-1158-1 · Cited by 48
Predicting future mooring line tension of floating structure by machine learning
Yang, Y., Peng, T., & Liao, S. (2022) · Ocean Engineering · doi:10.1016/j.oceaneng.2022.113470 · Cited by 40
“…been demonstrated that a ML model known as reservoir computing (RC) had great performance in modeling and predicting many complex dynamical systems (Jaeger and Haas, 2004; Pathak et al., 2017; Lu et al., 2018; Nakai and Saiki, 2018; Pathak et al., 2018; Jiang and Lai, 2019; Tanaka et al., 2019;…”
BackgroundMethodPhysical Reservoir Computing Using Tellurium-Based Gate-Tunable Artificial Photonic Synapses
Jo, H., Jang, J., Park, H. J., Lee, H., An, S. J., Hong, J. P., Jeong, M. S., & Oh, H. (2024) · ACS Nano · doi:10.1021/acsnano.4c10489 · Cited by 40
Reservoir Computing with Stochastic Bitstream Neurons
Verstraeten, D., Schrauwen, B., & Stroobandt, D. (2005) · Ghent University Academic Bibliography (Ghent University) · Cited by 37
Neuromorphic computing through time-multiplexing with a spin-torque nano-oscillator
Riou, M., Araujo, F. A., Torrejón, J., Tsunegi, S., Khalsa, G., Querlioz, D., Bortolotti, P., Cros, V., Yakushiji, K., Fukushima, A., et al. (2017) · Technical digest - International Electron Devices Meeting/Technical digest · doi:10.1109/iedm.2017.8268505 · Cited by 38
“1) [1], [2].”
Evolving Local Plasticity Rules for Synergistic Learning in Echo State Networks
Wang, X., Jin, Y., & Hao, K. (2019) · IEEE Transactions on Neural Networks and Learning Systems · doi:10.1109/tnnls.2019.2919903 · Cited by 43
“IS is essential to adapt the input data to the appropriate range so that the neurons can be properly activated [38].”
BackgroundResultTian, Z. (2020) · Journal of Ambient Intelligence and Humanized Computing · doi:10.1007/s12652-020-01920-4 · Cited by 38
Experimental realization of dual task processing with a photonic reservoir computer
Vatin, J., Rontani, D., & Sciamanna, M. (2020) · APL Photonics · doi:10.1063/5.0017574 · Cited by 41
An Echo State Network With Improved Topology for Time Series Prediction
Li, X., Bi, F., Yang, X. C., & Bi, X. (2022) · IEEE Sensors Journal · doi:10.1109/jsen.2022.3148742 · Cited by 39
A Time Series Modeling and Prediction of Wireless Network Traffic
Gowrishankar, S., & Satyanarayana, P. (2008) · International Journal of Interactive Mobile Technologies (iJIM) · doi:10.3991/ijim.v3i1.284 · Cited by 37
“Radial Basis Function (RBF) has achieved considerable success in nonlinear function prediction but the performance of RBF is less satisfactory for the nonlinear and nonstationary function prediction [28].”
Background
Metadata
- Authors
- Jaeger, H., & Haas, H.
- Affiliations
- Constructor University (DE)
- Corresponding author
- Jaeger, Herbert
- Venue
- Science
- Published
- 2004-04-02
- Volume
- 304
- Issue
- 5667
- Pages
- 78-80
- ISSN
- 0036-8075 (ISSN-L)
- 1095-9203
- OpenAlex
- W2118706537
- Cited by
- 3,811
- References
- 18
- Citations by year
- 2026: 257 · 2025: 314 · 2024: 315 · 2023: 332 · 2022: 300 · 2021: 355 · 2020: 259 · 2019: 309 · 2018: 237 · 2017: 160 · 2016: 142 · 2015: 128 · 2014: 120 · 2013: 107 · 2012: 100
- FWCI
- 33.93
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Science (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Reservoir Computing
- Physical Sciences › Computer Science › Artificial Intelligence › Neural Networks and Applications
- Physical Sciences › Physics and Astronomy › Statistical and Nonlinear Physics › Chaos control and synchronization
- Keywords
- echo state networks, chaotic time series prediction, recurrent neural networks, channel equalization, wireless communication, energy efficiency, reservoir computing
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.