An ensemble of five pruned two-class differentiators rejects over 90% of targeted and non-targeted misclassification attacks in transfer learning with less than 10% accuracy loss, under a black-box attack model.
Bunescu, and Gordon Stewart
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
baseline 1
citation-polarity summary
fields
cs.LG 1years
2019 1verdicts
CONDITIONAL 1roles
baseline 1polarities
baseline 1representative citing papers
citing papers explorer
-
Defeating Misclassification Attacks Against Transfer Learning
An ensemble of five pruned two-class differentiators rejects over 90% of targeted and non-targeted misclassification attacks in transfer learning with less than 10% accuracy loss, under a black-box attack model.