Amplifying Membership Exposure via Data Poisoning. (arXiv:2211.00463v1 [cs.CR])

As in-the-wild data are increasingly involved in the training stage, machine
learning applications become more susceptible to data poisoning attacks. Such
attacks typically lead to test-time accuracy degradation or controlled
misprediction. In this paper, we investigate the third type of exploitation of
data poisoning - increasing the risks of privacy leakage of benign training
samples. To this end, we demonstrate a set of data poisoning attacks to amplify
the membership exposure of the targeted class. We first propose a generic …

data data poisoning exposure poisoning