U-TraCE: a conformal prediction approach to uncertainty quantification in black-box models
Modern machine learning (ML) systems, particularly deep neural networks, have become increasingly complex and opaque, making reliable uncertainty quantification (UQ) essential. Conventional UQ methods often prove inadequate for black-box model evaluation scenarios, either because model internals are...
Zapisane w:
| Główni autorzy: | , |
|---|---|
| Format: | Artigo |
| Język: | Inglês |
| Wydane: |
IOP Publishing
2026-01-01
|
| Seria: | Machine Learning: Science and Technology |
| Hasła przedmiotowe: | |
| Dostęp online: | https://doi.org/10.1088/2632-2153/ae35ce |
| Etykiety: |
Nie ma etykietki, Dołącz pierwszą etykiete!
|