Towards Poisoning Fair Representations

arXiv:2309.16487v2 Announce Type: replace
Abstract: Fair machine learning seeks to mitigate model prediction bias against certain demographic subgroups such as elder and female. Recently, fair representation learning (FRL) trained by deep neural networks has demonstrated superior performance, whereby representations containing no demographic information are inferred from the data and then used as the input to classification or other downstream tasks. Despite the development of FRL methods, their vulnerability under data poisoning attack, a popular protocol to benchmark model robustness under …

abstract arxiv bias classification cs.lg data fair fair representations information machine machine learning networks neural networks performance prediction representation representation learning subgroups type