An adversarial Siamese network learns network-invariant attack representations and uses a ranking loss to retrieve more malicious events in transferred cyber detection models than naive transfer or CORAL on two public datasets.
Return of frustratingly easy domain adaptation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Semisupervised Adversarial Neural Networks for Cyber Security Transfer Learning
An adversarial Siamese network learns network-invariant attack representations and uses a ranking loss to retrieve more malicious events in transferred cyber detection models than naive transfer or CORAL on two public datasets.