Activation functions enabling the addition of neurons and layers without altering outcomes

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: López-Ureña, Sergio
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908350931795968
author López-Ureña, Sergio
author_facet López-Ureña, Sergio
contents In this work, we propose activation functions for neuronal networks that are refinable and sum the identity. This new class of activation functions allows the insertion of new layers between existing ones and/or the increase of neurons in a layer, both without altering the network outputs. Our approach is grounded in subdivision theory. The proposed activation functions are constructed from basic limit functions of convergent subdivision schemes. As a showcase of our results, we introduce a family of spline activation functions and provide comprehensive details for their practical implementation.
format Preprint
id arxiv_https___arxiv_org_abs_2410_12625
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Activation functions enabling the addition of neurons and layers without altering outcomes
López-Ureña, Sergio
Numerical Analysis
In this work, we propose activation functions for neuronal networks that are refinable and sum the identity. This new class of activation functions allows the insertion of new layers between existing ones and/or the increase of neurons in a layer, both without altering the network outputs. Our approach is grounded in subdivision theory. The proposed activation functions are constructed from basic limit functions of convergent subdivision schemes. As a showcase of our results, we introduce a family of spline activation functions and provide comprehensive details for their practical implementation.
title Activation functions enabling the addition of neurons and layers without altering outcomes
topic Numerical Analysis
url https://arxiv.org/abs/2410.12625