Stochastic dynamics learning with state-space systems
Not on the map. Why?
Abstract (arXiv)
This work advances the theoretical foundations of reservoir computing (RC) by providing a unified treatment of fading memory and the echo state property (ESP) in both deterministic and stochastic settings. We investigate state-space systems, a central model class in time series learning, and establish that fading memory and solution stability hold generically -- even in the absence of the ESP -- offering a robust explanation for the empirical success of RC models without strict contractivity conditions. In the stochastic case, we critically assess stochastic echo states, proposing a novel distributional perspective rooted in attractor dynamics on the space of probability distributions, which leads to a rich and coherent theory. Our results extend and generalize previous work on non-autonomous dynamical systems, offering new insights into causality, stability, and memory in RC models. This lays the groundwork for reliable generative modeling of temporal data in both deterministic and stochastic regimes.
Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 23 of its 89 references that are in the map. First posted on arXiv 11 Aug 2025.
Builds on
23 of 89 references.
References · 23
- 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,718
- 2014Deep Learning in Neural Networks: An Overviewcited by 770
- 2013Generating Sequences With Recurrent Neural Networkscited by 554
- 2020Time Series Forecasting With Deep Learning: A Surveycited by 459
- 2020Time Series Data Augmentation for Deep Learning: A Surveycited by 226
- 2019Recurrent Neural Networks for Time Series Forecasting: Current Status and Future Directionscited by 172
- 2023Machine Learning for Synthetic Data Generation: A Reviewcited by 120
- 2018Echo state networks are universalcited by 87
- 2020Physical reservoir computing -- An introductory perspectivecited by 58
- 2020Approximation Bounds for Random Neural Networks and Reservoir Systemscited by 52
- 2014On Particle Methods for Parameter Estimation in State-Space Modelscited by 47
- 2017Universal discrete-time reservoir computers with stochastic inputs and linear readouts using non-homogeneous state-affine systemscited by 39
11 more
- 2020Fading memory echo state networks are universalcited by 32
- 2020Combining Machine Learning with Knowledge-Based Modeling for Scalable Forecasting and Subgrid-Scale Closure of Large, Complex, Spatiotemporal Systemscited by 23
- 2019Differentiable reservoir computingcited by 20
- 2021Learning strange attractors with reservoir systemscited by 17
- 2023A Brief Survey on the Approximation Theory for Sequence Modellingcited by 15
- 2020Dimension reduction in recurrent networks by canonicalizationcited by 11
- 2023Optimal transport and Wasserstein distances for causal modelscited by 9
- 2021Embedding Information onto a Dynamical Systemcited by 5
- 2022Transport in reservoir computingcited by 5
- 2024State-space systems as dynamic generative modelscited by 3
- 2024Fading memory and the convolution theoremcited by 2
Built on by
Citing papers per year: 2025: 2; 2026: 1.
3 papers.