An analysis on the use of autoencoders for representation learning: Fundamentals, learning task case studies, explainability and challenges

Versión depositada: Manuscrito del autor (AAM / Post-print). Cumplimiento de política editorial: Según la política de autoarchivo de Elsevier para la revista Neurocomputing (ISSN: 0925-2312), se permite el depósito del manuscrito en el repositorio institucional del autor. El periodo de embargo de 24...

Descripción completa

Detalles Bibliográficos
Autores: Charte, David, Charte, Francisco, Del Jesus, María José, Herrera, Francisco
Tipo de recurso: artículo
Estado:Versión aceptada para publicación
Fecha de publicación:2020
País:España
Institución:Universidad de Jaén
Repositorio:RUJA. Repositorio Institucional de la Producción Científica de la Universidad de Jaén
OAI Identifier:oai:ruja.ujaen.es:10953/7131
Acceso en línea:https://www.sciencedirect.com/science/article/pii/S092523122030624X
https://hdl.handle.net/10953/7131
Access Level:acceso abierto
Palabra clave:Representation learning
Autoenconders
Deep learning
Feature extraction
004
004.3
004.6
004.8
id ES_97aec00ce765883fea12ab90d695357b
oai_identifier_str oai:ruja.ujaen.es:10953/7131
network_acronym_str ES
network_name_str España
repository_id_str
spelling An analysis on the use of autoencoders for representation learning: Fundamentals, learning task case studies, explainability and challengesCharte, DavidCharte, FranciscoDel Jesus, María JoséHerrera, FranciscoRepresentation learningAutoencondersDeep learningFeature extraction004004.3004.6004.8Versión depositada: Manuscrito del autor (AAM / Post-print). Cumplimiento de política editorial: Según la política de autoarchivo de Elsevier para la revista Neurocomputing (ISSN: 0925-2312), se permite el depósito del manuscrito en el repositorio institucional del autor. El periodo de embargo de 24 meses ha expirado (fecha de publicación original: 2020). Licencia aplicada: Este manuscrito se distribuye bajo una licencia Creative Commons Atribución-NoComercial-SinDerivadas 4.0 Internacional (CC-BY-NC-ND 4.0). Enlace a la versión oficial: El documento definitivo con formato de editorial está disponible en https://doi.org/10.1016/j.neucom.2020.04.057.In many machine learning tasks, learning a good representation of the data can be the key to building a well-performant solution. This is because most learning algorithms operate with the features in order to nd models for the data. For instance, classi cation performance can improve if the data is mapped to a space where classes are easily separated, and regression can be facilitated by finding a manifold of data in the feature space. As a general rule, features are transformed by means of statistical methods such as principal component analysis, or manifold learning techniques such as Isomap or locally linear embedding. From a plethora of representation learning methods, one of the most versatile tools is the autoencoder. In this paper we aim to demonstrate how to infuence its learned representations to achieve the desired learning behavior. To this end, we present a series of learning tasks: data embedding for visualization, image denoising, semantic hashing, detection of abnormal behaviors and instance generation. We model them from the representation learning perspective, following the state of the art methodologies in each eld. A solution is proposed for each task employing autoencoders as the only learning method. The theoretical developments are put into practice using a selection of datasets for the diferent problems and implementing each solution, followed by a discussion of the results in each case study and a brief explanation of other six learning applications. We also explore the current challenges and approaches to explainability in the context of autoencoders. All of this helps conclude that, thanks to alterations in their structure as well as their objective function, autoencoders may be the core of a possible solution to many problems which can be modeled as a transformation of the feature space.Elsevier202620262020info:eu-repo/semantics/articleinfo:eu-repo/semantics/acceptedVersionapplication/pdfhttps://www.sciencedirect.com/science/article/pii/S092523122030624Xhttps://hdl.handle.net/10953/7131reponame:RUJA. Repositorio Institucional de la Producción Científica de la Universidad de Jaéninstname:Universidad de JaénInglésNeurocomputingAttribution-NonCommercial-NoDerivs 3.0 Spainhttp://creativecommons.org/licenses/by-nc-nd/3.0/es/info:eu-repo/semantics/openAccessoai:ruja.ujaen.es:10953/71312026-06-24T12:41:07Z
dc.title.none.fl_str_mv An analysis on the use of autoencoders for representation learning: Fundamentals, learning task case studies, explainability and challenges
title An analysis on the use of autoencoders for representation learning: Fundamentals, learning task case studies, explainability and challenges
spellingShingle An analysis on the use of autoencoders for representation learning: Fundamentals, learning task case studies, explainability and challenges
Charte, David
Representation learning
Autoenconders
Deep learning
Feature extraction
004
004.3
004.6
004.8
title_short An analysis on the use of autoencoders for representation learning: Fundamentals, learning task case studies, explainability and challenges
title_full An analysis on the use of autoencoders for representation learning: Fundamentals, learning task case studies, explainability and challenges
title_fullStr An analysis on the use of autoencoders for representation learning: Fundamentals, learning task case studies, explainability and challenges
title_full_unstemmed An analysis on the use of autoencoders for representation learning: Fundamentals, learning task case studies, explainability and challenges
title_sort An analysis on the use of autoencoders for representation learning: Fundamentals, learning task case studies, explainability and challenges
dc.creator.none.fl_str_mv Charte, David
Charte, Francisco
Del Jesus, María José
Herrera, Francisco
author Charte, David
author_facet Charte, David
Charte, Francisco
Del Jesus, María José
Herrera, Francisco
author_role author
author2 Charte, Francisco
Del Jesus, María José
Herrera, Francisco
author2_role author
author
author
dc.subject.none.fl_str_mv Representation learning
Autoenconders
Deep learning
Feature extraction
004
004.3
004.6
004.8
topic Representation learning
Autoenconders
Deep learning
Feature extraction
004
004.3
004.6
004.8
description Versión depositada: Manuscrito del autor (AAM / Post-print). Cumplimiento de política editorial: Según la política de autoarchivo de Elsevier para la revista Neurocomputing (ISSN: 0925-2312), se permite el depósito del manuscrito en el repositorio institucional del autor. El periodo de embargo de 24 meses ha expirado (fecha de publicación original: 2020). Licencia aplicada: Este manuscrito se distribuye bajo una licencia Creative Commons Atribución-NoComercial-SinDerivadas 4.0 Internacional (CC-BY-NC-ND 4.0). Enlace a la versión oficial: El documento definitivo con formato de editorial está disponible en https://doi.org/10.1016/j.neucom.2020.04.057.
publishDate 2020
dc.date.none.fl_str_mv 2020
2026
2026
dc.type.none.fl_str_mv info:eu-repo/semantics/article
info:eu-repo/semantics/acceptedVersion
format article
status_str acceptedVersion
dc.identifier.none.fl_str_mv https://www.sciencedirect.com/science/article/pii/S092523122030624X
https://hdl.handle.net/10953/7131
url https://www.sciencedirect.com/science/article/pii/S092523122030624X
https://hdl.handle.net/10953/7131
dc.language.none.fl_str_mv Inglés
language_invalid_str_mv Inglés
dc.relation.none.fl_str_mv Neurocomputing
dc.rights.none.fl_str_mv Attribution-NonCommercial-NoDerivs 3.0 Spain
http://creativecommons.org/licenses/by-nc-nd/3.0/es/
info:eu-repo/semantics/openAccess
rights_invalid_str_mv Attribution-NonCommercial-NoDerivs 3.0 Spain
http://creativecommons.org/licenses/by-nc-nd/3.0/es/
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Elsevier
publisher.none.fl_str_mv Elsevier
dc.source.none.fl_str_mv reponame:RUJA. Repositorio Institucional de la Producción Científica de la Universidad de Jaén
instname:Universidad de Jaén
instname_str Universidad de Jaén
reponame_str RUJA. Repositorio Institucional de la Producción Científica de la Universidad de Jaén
collection RUJA. Repositorio Institucional de la Producción Científica de la Universidad de Jaén
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869414099498041344
score 15,811543

Similares en LA Referencia