Désapprentissage automatique


Révision datée du 6 mai 2024 à 08:34 par Pitpitt (discussion | contributions) (Page créée avec « == Définition == XXXXXXXXX == Français == ''' XXXXXXXXX ''' == Anglais == ''' Machine Unlearning ''' As our ML models today become larger and their (pre-)training sets grow to inscrutable sizes, people are increasingly interested in the concept of machine unlearning to edit away undesired things like private data, stale knowledge, copyrighted materials, toxic/unsafe content, dangerous capabilities, and misinformation, without retraining models from sc... »)
(diff) ← Version précédente | Voir la version actuelle (diff) | Version suivante → (diff)

Définition

XXXXXXXXX

Français

XXXXXXXXX

Anglais

Machine Unlearning

As our ML models today become larger and their (pre-)training sets grow to inscrutable sizes, people are increasingly interested in the concept of machine unlearning to edit away undesired things like private data, stale knowledge, copyrighted materials, toxic/unsafe content, dangerous capabilities, and misinformation, without retraining models from scratch.

Source

Source : deepgram

Source : research.google

Récupérée de « https://datafranca.org/wiki/index.php?title=Désapprentissage_automatique&oldid=103000 »

Contributeurs: Patrick Drouin, wiki