Approximation power of random neural networks
Abstract
This paper investigates the approximation power of three types of random neural networks: (a) infinite width networks, with weights following an arbitrary distribution; (b) finite width networks obtained by subsampling the preceding infinite width networks; (c) finite width networks obtained by starting with standard Gaussian initialization, and then adding a vanishingly small correction to the weights. The primary result is a fully quantified bound on the rate of approximation of general general continuous functions: in all three cases, a function can be approximated with complexity , where depends on continuity properties of and the complexity measure depends on the weight magnitudes and/or cardinalities. Along the way, a variety of ancillary results are developed: an exact construction of Gaussian densities with infinite width networks, an elementary stand-alone proof scheme for approximation via convolutions of radial basis functions, subsampling rates for infinite width networks, and depth separation for corrected networks.
Cite
@article{arxiv.1906.07709,
title = {Approximation power of random neural networks},
author = {Bolton Bailey and Ziwei Ji and Matus Telgarsky and Ruicheng Xian},
journal= {arXiv preprint arXiv:1906.07709},
year = {2019}
}
Comments
This submission constitutes a poor approach to the problem, and has no scientific purpose. A superior (different) approach (and stronger final result, also treating the NTK) has appeared in arXiv:1910.06956 ; please see that work instead