Can Triplet Loss Be Used for Multi-Label Few-Shot Classification? A Case Study
Few-shot learning is a deep learning subfield that is the focus of research nowadays. This paper addresses the research question of whether a triplet-trained Siamese network, initially designed for multi-class classification, can effectively handle multi-label classification. We conducted a case stu...
Elmentve itt :
| Szerzők: | |
|---|---|
| Dokumentumtípus: | Cikk |
| Megjelent: |
2023
|
| Sorozat: | INFORMATION (BASEL)
14 No. 10 |
| Tárgyszavak: | |
| doi: | 10.3390/info14100520 |
| mtmt: | 34154749 |
| Online Access: | http://publicatio.bibl.u-szeged.hu/28332 |
| LEADER | 02035nab a2200289 i 4500 | ||
|---|---|---|---|
| 001 | publ28332 | ||
| 005 | 20230926104750.0 | ||
| 008 | 230926s2023 hu o 0|| Angol d | ||
| 022 | |a 2078-2489 | ||
| 024 | 7 | |a 10.3390/info14100520 |2 doi | |
| 024 | 7 | |a 34154749 |2 mtmt | |
| 040 | |a SZTE Publicatio Repozitórium |b hun | ||
| 041 | |a Angol | ||
| 100 | 1 | |a Csányi Gergely | |
| 245 | 1 | 0 | |a Can Triplet Loss Be Used for Multi-Label Few-Shot Classification? A Case Study |h [elektronikus dokumentum] / |c Csányi Gergely |
| 260 | |c 2023 | ||
| 300 | |a 17 | ||
| 490 | 0 | |a INFORMATION (BASEL) |v 14 No. 10 | |
| 520 | 3 | |a Few-shot learning is a deep learning subfield that is the focus of research nowadays. This paper addresses the research question of whether a triplet-trained Siamese network, initially designed for multi-class classification, can effectively handle multi-label classification. We conducted a case study to identify any limitations in its application. The experiments were conducted on a dataset containing Hungarian legal decisions of administrative agencies in tax matters belonging to a major legal content provider. We also tested how different Siamese embeddings compare on classifying a previously non-existing label on a binary and a multi-label setting. We found that triplet-trained Siamese networks can be applied to perform classification but with a sampling restriction during training. We also found that the overlap between labels affects the results negatively. The few-shot model, seeing only ten examples for each label, provided competitive results compared to models trained on tens of thousands of court decisions using tf-idf vectorization and logistic regression. | |
| 650 | 4 | |a Egyéb természettudományok | |
| 700 | 0 | 1 | |a Vági Renátó |e aut |
| 700 | 0 | 1 | |a Megyeri Andrea |e aut |
| 700 | 0 | 1 | |a Fülöp Anna |e aut |
| 700 | 0 | 1 | |a Nagy Dániel |e aut |
| 700 | 0 | 1 | |a Vadász János Pál |e aut |
| 700 | 0 | 1 | |a Üveges István |e aut |
| 856 | 4 | 0 | |u http://publicatio.bibl.u-szeged.hu/28332/1/information-14-00520.pdf |z Dokumentum-elérés |