Universal Approximation Theorem for Neural Networks
This provides a foundational theoretical guarantee for neural network approximation ability, but it is an incremental explanation of existing theorems.
The paper explains the Universal Approximation Theorem for feedforward neural networks, which guarantees that neural networks can approximate any function in a certain space, and discusses the approximation rate problem relating the number of units to error.
Is there any theoretical guarantee for the approximation ability of neural networks? The answer to this question is the "Universal Approximation Theorem for Neural Networks". This theorem states that a neural network is dense in a certain function space under an appropriate setting. This paper is a comprehensive explanation of the universal approximation theorem for feedforward neural networks, its approximation rate problem (the relation between the number of intermediate units and the approximation error), and Barron space in Japanese.