Pith. sign in

Approximation Bounds for Random Neural Networks and Reservoir Systems

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it
abstract

This work studies approximation based on single-hidden-layer feedforward and recurrent neural networks with randomly generated internal weights. These methods, in which only the last layer of weights and a few hyperparameters are optimized, have been successfully applied in a wide range of static and dynamic learning problems. Despite the popularity of this approach in empirical tasks, important theoretical questions regarding the relation between the unknown function, the weight distribution, and the approximation rate have remained open. In this work it is proved that, as long as the unknown function, functional, or dynamical system is sufficiently regular, it is possible to draw the internal weights of the random (recurrent) neural network from a generic distribution (not depending on the unknown object) and quantify the error in terms of the number of neurons and the hyperparameters. In particular, this proves that echo state networks with randomly generated weights are capable of approximating a wide class of dynamical systems arbitrarily well and thus provides the first mathematical explanation for their empirically observed success at learning dynamical systems.

fields

nlin.CD 1

years

2019 1

verdicts

CONDITIONAL 1

representative citing papers

Embedding and Approximation Theorems for Echo State Networks

nlin.CD · 2019-08-14 · conditional · novelty 7.0

Under contraction conditions, an ESN trained on one-dimensional observations of an invertible dynamical system induces a C1 embedding with positive probability, and a linear readout makes the autonomous ESN topologically conjugate to the original system.

citing papers explorer

Showing 1 of 1 citing paper.

  • Embedding and Approximation Theorems for Echo State Networks nlin.CD · 2019-08-14 · conditional · none · ref 75 · internal anchor

    Under contraction conditions, an ESN trained on one-dimensional observations of an invertible dynamical system induces a C1 embedding with positive probability, and a linear readout makes the autonomous ESN topologically conjugate to the original system.