Embeddings between Barron spaces with higher order activation functions
share
The approximation properties of infinitely wide shallow neural networks heavily depend on the choice of the activation function. To understand this influence, we study embeddings between Barron spaces with different activation functions. These embeddings are proven by providing push-forward maps on the measures μ used to represent functions f. An activation function of particular interest is the rectified power unit (RePU) given by RePU_s(x)=max(0,x)^s. For many commonly used activation functions, the well-known Taylor remainder theorem can be used to construct a push-forward map, which allows us to prove the embedding of the associated Barron space into a Barron space with a RePU as activation function. Moreover, the Barron spaces associated with the RePU_s have a hierarchical structure similar to the Sobolev spaces H^m.
READ FULL TEXT