Stacked Ensemble Deep Random Vector Functional Link Network With Residual Learning for Medium-Scale Time-Series Forecasting

<p dir="ltr">The deep random vector functional link (dRVFL) and ensemble dRVFL (edRVFL) succeed in various tasks and achieve state-of-the-art performance compared with other randomized neural networks (NNs). However, existing edRVFL structures need more diversity and error correction...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Ruobin Gao (16003195) (author)
مؤلفون آخرون: Minghui Hu (2457952) (author), Ruilin Li (5627456) (author), Xuewen Luo (10012781) (author), Ponnuthurai Nagaratnam Suganthan (11274636) (author), M. Tanveer (1758181) (author)
منشور في: 2025
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513538488270848
author Ruobin Gao (16003195)
author2 Minghui Hu (2457952)
Ruilin Li (5627456)
Xuewen Luo (10012781)
Ponnuthurai Nagaratnam Suganthan (11274636)
M. Tanveer (1758181)
author2_role author
author
author
author
author
author_facet Ruobin Gao (16003195)
Minghui Hu (2457952)
Ruilin Li (5627456)
Xuewen Luo (10012781)
Ponnuthurai Nagaratnam Suganthan (11274636)
M. Tanveer (1758181)
author_role author
dc.creator.none.fl_str_mv Ruobin Gao (16003195)
Minghui Hu (2457952)
Ruilin Li (5627456)
Xuewen Luo (10012781)
Ponnuthurai Nagaratnam Suganthan (11274636)
M. Tanveer (1758181)
dc.date.none.fl_str_mv 2025-02-11T09:00:00Z
dc.identifier.none.fl_str_mv 10.1109/tnnls.2025.3529219
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/Stacked_Ensemble_Deep_Random_Vector_Functional_Link_Network_With_Residual_Learning_for_Medium-Scale_Time-Series_Forecasting/30198277
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Information and computing sciences
Artificial intelligence
Machine learning
Ensemble deep learning
forecasting
machine learning
multiple output layers
random vector functional link (RVFL) neural networks (NNs)
randomized NNs
transformers
Artificial neural networks
Training
Feature extraction
Vectors
Forecasting
Transformers
Learning systems
Uncertainty
Stochastic processes
Computational modeling
dc.title.none.fl_str_mv Stacked Ensemble Deep Random Vector Functional Link Network With Residual Learning for Medium-Scale Time-Series Forecasting
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p dir="ltr">The deep random vector functional link (dRVFL) and ensemble dRVFL (edRVFL) succeed in various tasks and achieve state-of-the-art performance compared with other randomized neural networks (NNs). However, existing edRVFL structures need more diversity and error correction ability in an independent network. Our work fills the gap by combining stacked deep blocks and residual learning with the edRVFL. Subsequently, we propose a novel dRVFL combined with residual learning, ResdRVFL, whose deep layers calibrate the wrong estimations from shallow layers. Additionally, we propose incorporating a scaling parameter to control the scaling of residuals from shallow layers, thus mitigating the risk of overfitting. Finally, we present an ensemble deep stacking network, SResdRVFL, based on ResdRVFL. SResdRVFL aggregates multiple blocks into a cohesive network, leveraging the benefits of deep learning and ensemble learning. We evaluate the proposed model on 28 datasets and compare it with the state-of-the-art methods. The comparative study demonstrates that the SResdRVFL is the best-performing approach in terms of average ranking and errors based on 28 datasets.</p><h2>Other Information</h2><p dir="ltr">Published in: IEEE Transactions on Neural Networks and Learning Systems<br>License: <a href="https://creativecommons.org/licenses/by/4.0/deed.en" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/tnnls.2025.3529219" target="_blank">https://dx.doi.org/10.1109/tnnls.2025.3529219</a></p>
eu_rights_str_mv openAccess
id Manara2_3f76355a5c8308f0feab96ed0511efcf
identifier_str_mv 10.1109/tnnls.2025.3529219
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/30198277
publishDate 2025
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Stacked Ensemble Deep Random Vector Functional Link Network With Residual Learning for Medium-Scale Time-Series ForecastingRuobin Gao (16003195)Minghui Hu (2457952)Ruilin Li (5627456)Xuewen Luo (10012781)Ponnuthurai Nagaratnam Suganthan (11274636)M. Tanveer (1758181)Information and computing sciencesArtificial intelligenceMachine learningEnsemble deep learningforecastingmachine learningmultiple output layersrandom vector functional link (RVFL) neural networks (NNs)randomized NNstransformersArtificial neural networksTrainingFeature extractionVectorsForecastingTransformersLearning systemsUncertaintyStochastic processesComputational modeling<p dir="ltr">The deep random vector functional link (dRVFL) and ensemble dRVFL (edRVFL) succeed in various tasks and achieve state-of-the-art performance compared with other randomized neural networks (NNs). However, existing edRVFL structures need more diversity and error correction ability in an independent network. Our work fills the gap by combining stacked deep blocks and residual learning with the edRVFL. Subsequently, we propose a novel dRVFL combined with residual learning, ResdRVFL, whose deep layers calibrate the wrong estimations from shallow layers. Additionally, we propose incorporating a scaling parameter to control the scaling of residuals from shallow layers, thus mitigating the risk of overfitting. Finally, we present an ensemble deep stacking network, SResdRVFL, based on ResdRVFL. SResdRVFL aggregates multiple blocks into a cohesive network, leveraging the benefits of deep learning and ensemble learning. We evaluate the proposed model on 28 datasets and compare it with the state-of-the-art methods. The comparative study demonstrates that the SResdRVFL is the best-performing approach in terms of average ranking and errors based on 28 datasets.</p><h2>Other Information</h2><p dir="ltr">Published in: IEEE Transactions on Neural Networks and Learning Systems<br>License: <a href="https://creativecommons.org/licenses/by/4.0/deed.en" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/tnnls.2025.3529219" target="_blank">https://dx.doi.org/10.1109/tnnls.2025.3529219</a></p>2025-02-11T09:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1109/tnnls.2025.3529219https://figshare.com/articles/journal_contribution/Stacked_Ensemble_Deep_Random_Vector_Functional_Link_Network_With_Residual_Learning_for_Medium-Scale_Time-Series_Forecasting/30198277CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/301982772025-02-11T09:00:00Z
spellingShingle Stacked Ensemble Deep Random Vector Functional Link Network With Residual Learning for Medium-Scale Time-Series Forecasting
Ruobin Gao (16003195)
Information and computing sciences
Artificial intelligence
Machine learning
Ensemble deep learning
forecasting
machine learning
multiple output layers
random vector functional link (RVFL) neural networks (NNs)
randomized NNs
transformers
Artificial neural networks
Training
Feature extraction
Vectors
Forecasting
Transformers
Learning systems
Uncertainty
Stochastic processes
Computational modeling
status_str publishedVersion
title Stacked Ensemble Deep Random Vector Functional Link Network With Residual Learning for Medium-Scale Time-Series Forecasting
title_full Stacked Ensemble Deep Random Vector Functional Link Network With Residual Learning for Medium-Scale Time-Series Forecasting
title_fullStr Stacked Ensemble Deep Random Vector Functional Link Network With Residual Learning for Medium-Scale Time-Series Forecasting
title_full_unstemmed Stacked Ensemble Deep Random Vector Functional Link Network With Residual Learning for Medium-Scale Time-Series Forecasting
title_short Stacked Ensemble Deep Random Vector Functional Link Network With Residual Learning for Medium-Scale Time-Series Forecasting
title_sort Stacked Ensemble Deep Random Vector Functional Link Network With Residual Learning for Medium-Scale Time-Series Forecasting
topic Information and computing sciences
Artificial intelligence
Machine learning
Ensemble deep learning
forecasting
machine learning
multiple output layers
random vector functional link (RVFL) neural networks (NNs)
randomized NNs
transformers
Artificial neural networks
Training
Feature extraction
Vectors
Forecasting
Transformers
Learning systems
Uncertainty
Stochastic processes
Computational modeling