Authors: Javid Ebrahimi,Anyi Rao,Daniel Lowd,Dejing Dou
Where published:
ACL 2018 7
ArXiv: 1712.06751
Document:
PDF
DOI
Artifact development version:
GitHub
Abstract URL: http://arxiv.org/abs/1712.06751v2
We propose an efficient method to generate white-box adversarial examples to
trick a character-level neural classifier. We find that only a few
manipulations are needed to greatly decrease the accuracy. Our method relies on
an atomic flip operation, which swaps one token for another, based on the
gradients of the one-hot input vectors. Due to efficiency of our method, we can
perform adversarial training which makes the model more robust to attacks at
test time. With the use of a few semantics-preserving constraints, we
demonstrate that HotFlip can be adapted to attack a word-level classifier as
well.