Balanced distributed augmentation for multi-label few shot learning with prototypical network

Basit Öğe Kaydı
dc.authoridAlper Öner / 0000-0002-6971-7494en_US
dc.authorscopusidAlper Öner / 50262728500en_US
dc.authorwosidAlper Öner / ACG-0850-2022
dc.contributor.authorMohammed, Hamza Haruna
dc.contributor.authorÖner, Alper
dc.date.accessioned2022-11-08T13:50:14Z
dc.date.available2022-11-08T13:50:14Z
dc.date.issued2022en_US
dc.departmentİstinye Üniversitesi, Mühendislik ve Doğa Bilimleri Fakültesi, Yazılım Mühendisliği Bölümüen_US
dc.description.abstractMany methods have been presented as a few shot learners in order to enhance few-shot learners. Some of these methods involve routine-based pre-trained language models and novel pipeline for automating the prompt generation. In this study, we propose a new evenly distributed data augmentation technique, which generates samples according to the probabilistic distribution of the relationship of each label with the mean of a label group. In the labeling phase, we present a semantic sentiment analysis approach in order to increase the realism of the data, in a more semantic augmentation way. The results show that this approach improves the few shot learners. In addition to this, we compare our adaptation approach to other traditional problem transformation methods. The newly developed approach outperforms these traditional methods, especially when the classifier learns from a limited number of samples.en_US
dc.identifier.citationMohammed, H. H., & Oner, A. (2022). Balanced distributed augmentation for multi-label few shot learning with prototypical network. Paper presented at the 2022 30th Signal Processing and Communications Applications Conference, SIU 2022, doi:10.1109/SIU55565.2022.9864875 Retrieved fromen_US
dc.identifier.doi10.1109/SIU55565.2022.9864875en_US
dc.identifier.scopus2-s2.0-85138704607en_US
dc.identifier.scopusqualityN/Aen_US
dc.identifier.urihttps://doi.org/10.1109/SIU55565.2022.9864875
dc.identifier.urihttps://hdl.handle.net/20.500.12713/3289
dc.indekslendigikaynakScopusen_US
dc.institutionauthorÖner, Alper
dc.language.isoenen_US
dc.publisherInstitute of Electrical and Electronics Engineers Inc.en_US
dc.relation.ispartofSignal Processing and Communications Applications Conferenceen_US
dc.relation.publicationcategoryMakale - Uluslararası Hakemli Dergi - Kurum Öğretim Elemanıen_US
dc.rightsinfo:eu-repo/semantics/closedAccessen_US
dc.subjectFew Shot Learningen_US
dc.subjectMeta-learningen_US
dc.subjectMulti-label Classificationen_US
dc.subjectPrototypical Networken_US
dc.subjectText Data Augmentationen_US
dc.titleBalanced distributed augmentation for multi-label few shot learning with prototypical networken_US
dc.typeArticleen_US

Dosyalar

Orijinal paket
Listeleniyor 1 - 1 / 1
Küçük Resim Yok
İsim:
Balanced_Distributed_Augmentation_for_Multi-Label_Few_Shot_Learning_with_Prototypical_Network.pdf
Boyut:
922.1 KB
Biçim:
Adobe Portable Document Format
Açıklama:
Tam Metin / Full Text
İndir
Lisans paketi
Listeleniyor 1 - 1 / 1
Küçük Resim Yok
İsim:
license.txt
Boyut:
1.44 KB
Biçim:
Item-specific license agreed upon to submission
Açıklama:
İndir

Koleksiyon

Scopus İndeksli Yayınlar Koleksiyonu