Challenges for the Repeatability of Deep Learning Models
Deep learning training typically starts with a random sampling initialization approach to set the weights of trainable layers. Therefore, different and/or uncontrolled weight initialization prevents learning the same model multiple times. Consequently, such models yield different results during test...
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Artigo |
| Langue: | Inglês |
| Publié: |
IEEE
2020-01-01
|
| Collection: | IEEE Access |
| Sujets: | |
| Accès en ligne: | https://ieeexplore.ieee.org/document/9266043/ |
| Tags: |
Pas de tags, Soyez le premier à ajouter un tag!
|