DPL++ : Advancing the Network Performance via Image and Label Perturbations
Recent advances in supervised learning have predominantly focused on regularizations, optimizers, and architectures, yet the potential of simultaneously optimizing data distributions and supervisory signals for training samples remains underexplored. In this paper, we propose a novel paradigm that l...
Ausführliche Beschreibung
Bibliographische Detailangaben
| Veröffentlicht in: | IEEE transactions on pattern analysis and machine intelligence. - 1979. - 47(2025), 11 vom: 06. Okt., Seite 10262-10279
|
|---|
| 1. Verfasser: |
Song, Zifan
(VerfasserIn) |
|---|
| Weitere Verfasser: |
Gong, Xiao,
Hu, Guosheng,
Dou, Shuguang,
Zhao, Qingsong,
Zhao, Cairong |
|---|
| Format: | Online-Aufsatz
|
|---|
| Sprache: | English |
|---|
| Veröffentlicht: |
2025
|
|---|
| Zugriff auf das übergeordnete Werk: | IEEE transactions on pattern analysis and machine intelligence
|
|---|
| Schlagworte: | Journal Article |
|---|