Arbitrary nonlinearity is sufficient to represent all functions by neural networks: A theorem

Explore this paper's citation graph

Summary

It is proved that if the authors have neurons implementing arbitrary linear functions and a neuron implementing one (arbitrary but smooth) nonlinear function g(x), then for every continuous function f(x1, …, xm) of arbitrarily many variables and for arbitrary e > 0 they can construct a network that consists of g-neurons and linear neurons and computes f with precision e.

Type
article
Published
1991-06-01
Cited by
166
References
10

Keywords

Artificial neural network, Nonlinear system, Construct (python library), Function (biology), Mathematics

References

Cited by

Related papers