UNIVERSALITY OF NEURAL NETWORKS WITH A SIGMOIDAL ACTIVATION OR DISCRIMINATORY FUNCTIONS ON FUNCTIONAL BANACH LATTICES OVER THE REAL LINE

The universality of neural networks with sigmoidal or discriminatory functions are characterized on functional Banach lattices. This study is an extension a previous similar problem unbounded exponent in variable Lebesgue spaces, which contain essentially bounded measurable functions. However, the notion lattices includes other fundamental function spaces and our result natural generalization well-known by Cybenko. We examine structure uniformly continuous for this purpose provide over domains.