Per-instance privacy losses, estimated from gradient norms during training, predict the number of fine-tuning steps needed for machine unlearning and rank data points by unlearning difficulty.
Unlearning time vs
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Leveraging Per-Instance Privacy for Machine Unlearning
Per-instance privacy losses, estimated from gradient norms during training, predict the number of fine-tuning steps needed for machine unlearning and rank data points by unlearning difficulty.