On the amplification of security and privacy risks by post-hoc explanations in machine learning models. (arXiv:2206.14004v1 [cs.LG])

A variety of explanation methods have been proposed in recent years to help
users gain insights into the results returned by neural networks, which are
otherwise complex and opaque black-boxes. However, explanations give rise to
potential side-channels that can be leveraged by an adversary for mounting
attacks on the system. In particular, post-hoc explanation methods that
highlight input dimensions according to their importance or relevance to the
result also leak information that weakens security and privacy. In this work,
we …

amplification lg machine machine learning machine learning models privacy security