A deep network's intermediate layers can be ranked by privacy risk using estimated degrees of freedom and Jacobian rank, which the authors link to membership inference attack success.
Enhanced membership inference attacks against machine learning models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CV 1years
2024 1verdicts
REJECT 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Intermediate Outputs Are More Sensitive Than You Think
A deep network's intermediate layers can be ranked by privacy risk using estimated degrees of freedom and Jacobian rank, which the authors link to membership inference attack success.