Maximal function pooling with applications
This addresses pooling function design in neural networks, offering a new option for researchers and practitioners, though it appears incremental as it builds on established pooling techniques.
The authors tackled the problem of pooling in neural networks by proposing maxfun pooling, a novel strategy inspired by the Hardy-Littlewood maximal function that interpolates between max and average pooling. They demonstrated its viability in convolutional sparse coding and image classification, showing it as a competitive alternative to existing methods.
Inspired by the Hardy-Littlewood maximal function, we propose a novel pooling strategy which is called maxfun pooling. It is presented both as a viable alternative to some of the most popular pooling functions, such as max pooling and average pooling, and as a way of interpolating between these two algorithms. We demonstrate the features of maxfun pooling with two applications: first in the context of convolutional sparse coding, and then for image classification.