INVESTIGATING THE IMPACT OF FEATURE SELECTION METHODS ON THE PERFORMANCE OF DEEP NEURAL NETWORKS FOR IMAGE RECOGNITION TASKS

Fuente: Zenodo
Guardado en:
Detalles Bibliográficos
Autor principal: Researcher
Formato: Recurso digital
Lenguaje:inglés
Publicado: Zenodo 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866902009754419200
author Researcher
author_facet Researcher
contents <p><span lang="EN-US">Feature selection has been a fundamental step in traditional machine learning pipelines, yet its role in deep learning remains underexplored. In this study, we investigate how different feature selection methods—both filter-based and embedded—affect the performance of deep neural networks (DNNs) on image recognition tasks. Using benchmark datasets such as CIFAR-10 and Fashion-MNIST, we apply popular feature selection techniques prior to training convolutional neural networks (CNNs). Our results show that while traditional feature selection often degrades performance when applied naively to image data, hybrid methods integrating domain knowledge and structural pruning yield notable improvements in training efficiency and generalization. This paper provides a nuanced view of how classical concepts like feature relevance intersect with the end-to-end learning paradigm of deep networks.</span></p>
format Recurso digital
id zenodo_https___doi_org_10_5281_zenodo_15559499
institution Zenodo
language eng
publishDate 2025
publisher Zenodo
record_format zenodo
spellingShingle INVESTIGATING THE IMPACT OF FEATURE SELECTION METHODS ON THE PERFORMANCE OF DEEP NEURAL NETWORKS FOR IMAGE RECOGNITION TASKS
Researcher
Feature selection, deep neural networks, image recognition, CNN, dimensionality reduction, filter methods, embedded methods.
<p><span lang="EN-US">Feature selection has been a fundamental step in traditional machine learning pipelines, yet its role in deep learning remains underexplored. In this study, we investigate how different feature selection methods—both filter-based and embedded—affect the performance of deep neural networks (DNNs) on image recognition tasks. Using benchmark datasets such as CIFAR-10 and Fashion-MNIST, we apply popular feature selection techniques prior to training convolutional neural networks (CNNs). Our results show that while traditional feature selection often degrades performance when applied naively to image data, hybrid methods integrating domain knowledge and structural pruning yield notable improvements in training efficiency and generalization. This paper provides a nuanced view of how classical concepts like feature relevance intersect with the end-to-end learning paradigm of deep networks.</span></p>
title INVESTIGATING THE IMPACT OF FEATURE SELECTION METHODS ON THE PERFORMANCE OF DEEP NEURAL NETWORKS FOR IMAGE RECOGNITION TASKS
topic Feature selection, deep neural networks, image recognition, CNN, dimensionality reduction, filter methods, embedded methods.
url https://doi.org/10.5281/zenodo.15559499