Simple reservoir computing capitalizing on the nonlinear response of materials: theory and physical implementations

S Kan, K Nakajima, Y Takeshima, T Asai… - Physical review applied, 2021 - APS
The potential of nonlinear dynamical systems serving as reservoirs has attracted much
attention for the physical realization of reservoir computing (RC). Here, we propose a …

Modularity and multitasking in neuro-memristive reservoir networks

A Loeffler, R Zhu, J Hochstetter… - Neuromorphic …, 2021 - iopscience.iop.org
The human brain seemingly effortlessly performs multiple concurrent and elaborate tasks in
response to complex, dynamic sensory input from our environment. This capability has been …

Universal discrete-time reservoir computers with stochastic inputs and linear readouts using non-homogeneous state-affine systems

L Grigoryeva, JP Ortega - Journal of Machine Learning Research, 2018 - jmlr.org
A new class of non-homogeneous state-affine systems is introduced for use in reservoir
computing. Sufficient conditions are identified that guarantee first, that the associated …

Risk bounds for reservoir computing

L Gonon, L Grigoryeva, JP Ortega - Journal of Machine Learning Research, 2020 - jmlr.org
We analyze the practices of reservoir computing in the framework of statistical learning
theory. In particular, we derive finite sample upper bounds for the generalization error …

Convolutional multitimescale echo state network

Q Ma, E Chen, Z Lin, J Yan, Z Yu… - IEEE Transactions on …, 2019 - ieeexplore.ieee.org
As efficient recurrent neural network (RNN) models, echo state networks (ESNs) have
attracted widespread attention and been applied in many application domains in the last …

Differentiable reservoir computing

L Grigoryeva, JP Ortega - Journal of Machine Learning Research, 2019 - jmlr.org
Numerous results in learning and approximation theory have evidenced the importance of
differentiability at the time of countering the curse of dimensionality. In the context of …

Memory of recurrent networks: Do we compute it right?

G Ballarin, L Grigoryeva, JP Ortega - Journal of Machine Learning …, 2024 - jmlr.org
Numerical evaluations of the memory capacity (MC) of recurrent neural networks reported in
the literature often contradict well-established theoretical bounds. In this paper, we study the …

Memory and forecasting capacities of nonlinear recurrent networks

L Gonon, L Grigoryeva, JP Ortega - Physica D: Nonlinear Phenomena, 2020 - Elsevier
The notion of memory capacity, originally introduced for echo state and linear networks with
independent inputs, is generalized to nonlinear recurrent networks with stationary but …

Short-wavelength reverberant wave systems for physical realization of reservoir computing

S Ma, TM Antonsen, SM Anlage, E Ott - Physical Review Research, 2022 - APS
Machine learning (ML) has found widespread application over a broad range of important
tasks. To enhance ML performance, researchers have investigated computational …

[HTML][HTML] Reservoir computing for macroeconomic forecasting with mixed-frequency data

G Ballarin, P Dellaportas, L Grigoryeva, M Hirt… - International Journal of …, 2024 - Elsevier
Macroeconomic forecasting has recently started embracing techniques that can deal with
large-scale datasets and series with unequal release periods. Mixed-data sampling (MIDAS) …