Shared-Weights Extender and Gradient Voting for Neural Network Expansion

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Chatzis, Nikolas, Kordonis, Ioannis, Theodosis, Manos, Maragos, Petros
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866908555282481152
author Chatzis, Nikolas
Kordonis, Ioannis
Theodosis, Manos
Maragos, Petros
author_facet Chatzis, Nikolas
Kordonis, Ioannis
Theodosis, Manos
Maragos, Petros
contents Expanding neural networks during training is a promising way to augment capacity without retraining larger models from scratch. However, newly added neurons often fail to adjust to a trained network and become inactive, providing no contribution to capacity growth. We propose the Shared-Weights Extender (SWE), a novel method explicitly designed to prevent inactivity of new neurons by coupling them with existing ones for smooth integration. In parallel, we introduce the Steepest Voting Distributor (SVoD), a gradient-based method for allocating neurons across layers during deep network expansion. Our extensive benchmarking on four datasets shows that our method can effectively suppress neuron inactivity and achieve better performance compared to other expanding methods and baselines.
format Preprint
id arxiv_https___arxiv_org_abs_2509_18842
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Shared-Weights Extender and Gradient Voting for Neural Network Expansion
Chatzis, Nikolas
Kordonis, Ioannis
Theodosis, Manos
Maragos, Petros
Machine Learning
Expanding neural networks during training is a promising way to augment capacity without retraining larger models from scratch. However, newly added neurons often fail to adjust to a trained network and become inactive, providing no contribution to capacity growth. We propose the Shared-Weights Extender (SWE), a novel method explicitly designed to prevent inactivity of new neurons by coupling them with existing ones for smooth integration. In parallel, we introduce the Steepest Voting Distributor (SVoD), a gradient-based method for allocating neurons across layers during deep network expansion. Our extensive benchmarking on four datasets shows that our method can effectively suppress neuron inactivity and achieve better performance compared to other expanding methods and baselines.
title Shared-Weights Extender and Gradient Voting for Neural Network Expansion
topic Machine Learning
url https://arxiv.org/abs/2509.18842