MetaAdvDet uses a MAML-style double-network meta-learner to detect evolving adversarial attacks with one to five labeled examples, outperforming non-meta baselines on most tested benchmarks.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
MetaAdvDet: Towards Robust Detection of Evolving Adversarial Attacks
MetaAdvDet uses a MAML-style double-network meta-learner to detect evolving adversarial attacks with one to five labeled examples, outperforming non-meta baselines on most tested benchmarks.