Empirical study of the modulus as activation function in computer vision applications
Abstract
In this work we propose a new non-monotonic activation function: the modulus. The majority of the reported research on nonlinearities is focused on monotonic functions. We empirically demonstrate how by using the modulus activation function on computer vision tasks the models generalize better than with other nonlinearities - up to a 15% accuracy increase in CIFAR100 and 4% in CIFAR10, relative to the best of the benchmark activations tested. With the proposed activation function the vanishing gradient and dying neurons problems disappear, because the derivative of the activation function is always 1 or -1. The simplicity of the proposed function and its derivative make this solution specially suitable for TinyML and hardware applications.
Keywords
Cite
@article{arxiv.2301.05993,
title = {Empirical study of the modulus as activation function in computer vision applications},
author = {Iván Vallés-Pérez and Emilio Soria-Olivas and Marcelino Martínez-Sober and Antonio J. Serrano-López and Joan Vila-Francés and Juan Gómez-Sanchís},
journal= {arXiv preprint arXiv:2301.05993},
year = {2023}
}
Comments
Accepted at Engineering Applications of AI