Euler State Networks: Non-dissipative Reservoir Computing
This addresses the need for efficient and effective reservoir computing in time-series analysis, though it is an incremental improvement over existing RC models.
The paper tackles the problem of designing stable and non-dissipative reservoir computing models for long-term memory tasks, resulting in a model that matches or exceeds the accuracy of trainable recurrent neural networks while achieving up to ≈490-fold savings in computation time and ≈1750-fold savings in energy consumption.
Inspired by the numerical solution of ordinary differential equations, in this paper we propose a novel Reservoir Computing (RC) model, called the Euler State Network (EuSN). The presented approach makes use of forward Euler discretization and antisymmetric recurrent matrices to design reservoir dynamics that are both stable and non-dissipative by construction. Our mathematical analysis shows that the resulting model is biased towards a unitary effective spectral radius and zero local Lyapunov exponents, intrinsically operating near to the edge of stability. Experiments on long-term memory tasks show the clear superiority of the proposed approach over standard RC models in problems requiring effective propagation of input information over multiple time-steps. Furthermore, results on time-series classification benchmarks indicate that EuSN is able to match (or even exceed) the accuracy of trainable Recurrent Neural Networks, while retaining the training efficiency of the RC family, resulting in up to $\approx$ 490-fold savings in computation time and $\approx$ 1750-fold savings in energy consumption.