Impact of input dataset size and fine-tuning on faster R-CNN with transfer learning

Deep learning models are widely used for machine learning tasks such as object detection. The lack of available data to train these models is a common hindrance in many industrial applications, where data gathering/annotation and insufficient computational resources often impose a barrier to the fin...

ver descrição completa

Detalhes bibliográficos
Autor: Zheng, Wei
Formato: tesis de maestría
Fecha de publicación:2023
País:España
Recursos:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2117/406884
Acesso em linha:https://hdl.handle.net/2117/406884
Access Level:acceso abierto
Palavra-chave:Deep learning
Neural networks (Computer science)
Aprenentatge profund
Xarxes neuronals (Informàtica)
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial
id ES_b036df879952797310f709eec70ff2bc
oai_identifier_str oai:upcommons.upc.edu:2117/406884
network_acronym_str ES
network_name_str España
repository_id_str
spelling Impact of input dataset size and fine-tuning on faster R-CNN with transfer learningZheng, WeiDeep learningNeural networks (Computer science)Aprenentatge profundXarxes neuronals (Informàtica)Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificialDeep learning models are widely used for machine learning tasks such as object detection. The lack of available data to train these models is a common hindrance in many industrial applications, where data gathering/annotation and insufficient computational resources often impose a barrier to the financial feasibility of deep learning implementations. Transfer learning is a possible answer to this challenge by exploiting the information learned by a model from data in a different domain than that of the target dataset. This technique has been typically applied on the backbone network of a two-stage object detection pipeline. In this work, we investigate the association between the input dataset size and the proportion of trainable layers in the backbone. In particular, we show some interesting findings on Faster R-CNN ResNet-50 FPN, a state-of-the-art object detection model, and MS COCO, a benchmarking dataset. The outcomes of our experiments indicate that, although a model generally performs better when trained with more layers fine-tuned to the training data, such an advantage reduces as the input dataset becomes smaller, as unfreezing too many layers can even lead to a severe overfitting problem. Choosing the right number of layers to freeze when applying transfer learning not only allows the model to reach its best possible performance but also saves computational resources and training time. Additionally, we explore the association between the effect of learning rate decay and the input dataset size, and also discuss the advantage of using pre-trained weights when compared to training a network from scratch.Universitat Politècnica de CatalunyaCabral Pinheiro, Victor HenriqueBjörklund, Tomas20232023-11-1520242024-04-22master thesishttp://purl.org/coar/resource_type/c_bdccNAhttp://purl.org/coar/version/c_be7fb7dd8ff6fe43info:eu-repo/semantics/masterThesisapplication/pdfhttps://hdl.handle.net/2117/406884reponame:UPCommons. Portal del coneixement obert de la UPCinstname:Universitat Politècnica de Catalunya (UPC)Inglésengopen accesshttp://purl.org/coar/access_right/c_abf2info:eu-repo/semantics/openAccessoai:upcommons.upc.edu:2117/4068842026-05-27T15:37:01Z
dc.title.none.fl_str_mv Impact of input dataset size and fine-tuning on faster R-CNN with transfer learning
title Impact of input dataset size and fine-tuning on faster R-CNN with transfer learning
spellingShingle Impact of input dataset size and fine-tuning on faster R-CNN with transfer learning
Zheng, Wei
Deep learning
Neural networks (Computer science)
Aprenentatge profund
Xarxes neuronals (Informàtica)
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial
title_short Impact of input dataset size and fine-tuning on faster R-CNN with transfer learning
title_full Impact of input dataset size and fine-tuning on faster R-CNN with transfer learning
title_fullStr Impact of input dataset size and fine-tuning on faster R-CNN with transfer learning
title_full_unstemmed Impact of input dataset size and fine-tuning on faster R-CNN with transfer learning
title_sort Impact of input dataset size and fine-tuning on faster R-CNN with transfer learning
dc.creator.none.fl_str_mv Zheng, Wei
author Zheng, Wei
author_facet Zheng, Wei
author_role author
dc.contributor.none.fl_str_mv Cabral Pinheiro, Victor Henrique
Björklund, Tomas
dc.subject.none.fl_str_mv Deep learning
Neural networks (Computer science)
Aprenentatge profund
Xarxes neuronals (Informàtica)
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial
topic Deep learning
Neural networks (Computer science)
Aprenentatge profund
Xarxes neuronals (Informàtica)
Àrees temàtiques de la UPC::Informàtica::Intel·ligència artificial
description Deep learning models are widely used for machine learning tasks such as object detection. The lack of available data to train these models is a common hindrance in many industrial applications, where data gathering/annotation and insufficient computational resources often impose a barrier to the financial feasibility of deep learning implementations. Transfer learning is a possible answer to this challenge by exploiting the information learned by a model from data in a different domain than that of the target dataset. This technique has been typically applied on the backbone network of a two-stage object detection pipeline. In this work, we investigate the association between the input dataset size and the proportion of trainable layers in the backbone. In particular, we show some interesting findings on Faster R-CNN ResNet-50 FPN, a state-of-the-art object detection model, and MS COCO, a benchmarking dataset. The outcomes of our experiments indicate that, although a model generally performs better when trained with more layers fine-tuned to the training data, such an advantage reduces as the input dataset becomes smaller, as unfreezing too many layers can even lead to a severe overfitting problem. Choosing the right number of layers to freeze when applying transfer learning not only allows the model to reach its best possible performance but also saves computational resources and training time. Additionally, we explore the association between the effect of learning rate decay and the input dataset size, and also discuss the advantage of using pre-trained weights when compared to training a network from scratch.
publishDate 2023
dc.date.none.fl_str_mv 2023
2023-11-15
2024
2024-04-22
dc.type.none.fl_str_mv master thesis
http://purl.org/coar/resource_type/c_bdcc
NA
http://purl.org/coar/version/c_be7fb7dd8ff6fe43
dc.type.openaire.fl_str_mv info:eu-repo/semantics/masterThesis
format masterThesis
dc.identifier.none.fl_str_mv https://hdl.handle.net/2117/406884
url https://hdl.handle.net/2117/406884
dc.language.none.fl_str_mv Inglés
eng
language_invalid_str_mv Inglés
language eng
dc.rights.none.fl_str_mv open access
http://purl.org/coar/access_right/c_abf2
dc.rights.openaire.fl_str_mv info:eu-repo/semantics/openAccess
rights_invalid_str_mv open access
http://purl.org/coar/access_right/c_abf2
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Universitat Politècnica de Catalunya
publisher.none.fl_str_mv Universitat Politècnica de Catalunya
dc.source.none.fl_str_mv reponame:UPCommons. Portal del coneixement obert de la UPC
instname:Universitat Politècnica de Catalunya (UPC)
instname_str Universitat Politècnica de Catalunya (UPC)
reponame_str UPCommons. Portal del coneixement obert de la UPC
collection UPCommons. Portal del coneixement obert de la UPC
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869416788620476416
score 15.198674

Registos relacionados