Active Learning Based Federated Learning for Waste and Natural Disaster Image Classification

<p>The feasibility of Federated Learning (FL) is highly dependent on the training and inference capabilities of local models, which are subject to the availability of meaningful and annotated data. The availability of such data is in turn contingent on the tedious and time-consuming annotation...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Lulwa Ahmed (16869936) (author)
مؤلفون آخرون: Kashif Ahmad (12592762) (author), Naina Said (16869939) (author), Basheer Qolomany (16855527) (author), Junaid Qadir (16494902) (author), Ala Al-Fuqaha (4434340) (author)
منشور في: 2020
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513561273827328
author Lulwa Ahmed (16869936)
author2 Kashif Ahmad (12592762)
Naina Said (16869939)
Basheer Qolomany (16855527)
Junaid Qadir (16494902)
Ala Al-Fuqaha (4434340)
author2_role author
author
author
author
author
author_facet Lulwa Ahmed (16869936)
Kashif Ahmad (12592762)
Naina Said (16869939)
Basheer Qolomany (16855527)
Junaid Qadir (16494902)
Ala Al-Fuqaha (4434340)
author_role author
dc.creator.none.fl_str_mv Lulwa Ahmed (16869936)
Kashif Ahmad (12592762)
Naina Said (16869939)
Basheer Qolomany (16855527)
Junaid Qadir (16494902)
Ala Al-Fuqaha (4434340)
dc.date.none.fl_str_mv 2020-11-17T00:00:00Z
dc.identifier.none.fl_str_mv 10.1109/access.2020.3038676
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/Active_Learning_Based_Federated_Learning_for_Waste_and_Natural_Disaster_Image_Classification/24015858
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
Data management and data science
Machine learning
Training
Training data
Manuals
Machine learning
Inspection
Collaborative work
Data models
Federated learning
Deep learning
Active learning
CNNs
LSTM
Natural disasters
Waste classification
dc.title.none.fl_str_mv Active Learning Based Federated Learning for Waste and Natural Disaster Image Classification
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p>The feasibility of Federated Learning (FL) is highly dependent on the training and inference capabilities of local models, which are subject to the availability of meaningful and annotated data. The availability of such data is in turn contingent on the tedious and time-consuming annotation job that typically requires the manual analysis of training samples. Active Learning (AL) provides an alternative solution allowing a Machine Learning (ML) model to automatically choose and label the data from which it learns without involving manual inspection of each training sample. In this work, we explore how FL can benefit from unlabelled data available at each participating client using AL. To this aim, we propose an AL-based FL framework by employing and evaluating several AL methods in two different application domains. Through an extensive experimentation setup, we show that AL is equally useful in federated and centralized learning by achieving comparable results with manually labeled data using fewer samples without involving human annotators in collecting training data. We also demonstrated that the proposed method is dataset/application independent by evaluating the proposed method in two interesting applications, namely natural disaster analysis and waste classification, having different properties and challenges. Promising results are obtained on both applications resulting in comparable results against the best-case scenario where each sample is manually analyzed and annotated (Baseline 1), and improvement of 3.1% and 4% with best methods respectively over the training sets with irrelevant images on natural disaster and waste classification datasets (Baseline 2).</p><h2>Other Information</h2><p>Published in: IEEE Access<br>License: <a href="https://creativecommons.org/licenses/by/4.0/legalcode" 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/access.2020.3038676" target="_blank">https://dx.doi.org/10.1109/access.2020.3038676</a></p>
eu_rights_str_mv openAccess
id Manara2_4310ccdf4b63e7b1062c5b71d1ae0a0f
identifier_str_mv 10.1109/access.2020.3038676
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/24015858
publishDate 2020
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling Active Learning Based Federated Learning for Waste and Natural Disaster Image ClassificationLulwa Ahmed (16869936)Kashif Ahmad (12592762)Naina Said (16869939)Basheer Qolomany (16855527)Junaid Qadir (16494902)Ala Al-Fuqaha (4434340)Information and computing sciencesData management and data scienceMachine learningTrainingTraining dataManualsMachine learningInspectionCollaborative workData modelsFederated learningDeep learningActive learningCNNsLSTMNatural disastersWaste classification<p>The feasibility of Federated Learning (FL) is highly dependent on the training and inference capabilities of local models, which are subject to the availability of meaningful and annotated data. The availability of such data is in turn contingent on the tedious and time-consuming annotation job that typically requires the manual analysis of training samples. Active Learning (AL) provides an alternative solution allowing a Machine Learning (ML) model to automatically choose and label the data from which it learns without involving manual inspection of each training sample. In this work, we explore how FL can benefit from unlabelled data available at each participating client using AL. To this aim, we propose an AL-based FL framework by employing and evaluating several AL methods in two different application domains. Through an extensive experimentation setup, we show that AL is equally useful in federated and centralized learning by achieving comparable results with manually labeled data using fewer samples without involving human annotators in collecting training data. We also demonstrated that the proposed method is dataset/application independent by evaluating the proposed method in two interesting applications, namely natural disaster analysis and waste classification, having different properties and challenges. Promising results are obtained on both applications resulting in comparable results against the best-case scenario where each sample is manually analyzed and annotated (Baseline 1), and improvement of 3.1% and 4% with best methods respectively over the training sets with irrelevant images on natural disaster and waste classification datasets (Baseline 2).</p><h2>Other Information</h2><p>Published in: IEEE Access<br>License: <a href="https://creativecommons.org/licenses/by/4.0/legalcode" 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/access.2020.3038676" target="_blank">https://dx.doi.org/10.1109/access.2020.3038676</a></p>2020-11-17T00:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1109/access.2020.3038676https://figshare.com/articles/journal_contribution/Active_Learning_Based_Federated_Learning_for_Waste_and_Natural_Disaster_Image_Classification/24015858CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/240158582020-11-17T00:00:00Z
spellingShingle Active Learning Based Federated Learning for Waste and Natural Disaster Image Classification
Lulwa Ahmed (16869936)
Information and computing sciences
Data management and data science
Machine learning
Training
Training data
Manuals
Machine learning
Inspection
Collaborative work
Data models
Federated learning
Deep learning
Active learning
CNNs
LSTM
Natural disasters
Waste classification
status_str publishedVersion
title Active Learning Based Federated Learning for Waste and Natural Disaster Image Classification
title_full Active Learning Based Federated Learning for Waste and Natural Disaster Image Classification
title_fullStr Active Learning Based Federated Learning for Waste and Natural Disaster Image Classification
title_full_unstemmed Active Learning Based Federated Learning for Waste and Natural Disaster Image Classification
title_short Active Learning Based Federated Learning for Waste and Natural Disaster Image Classification
title_sort Active Learning Based Federated Learning for Waste and Natural Disaster Image Classification
topic Information and computing sciences
Data management and data science
Machine learning
Training
Training data
Manuals
Machine learning
Inspection
Collaborative work
Data models
Federated learning
Deep learning
Active learning
CNNs
LSTM
Natural disasters
Waste classification