Let OPT denote an optimal offline algorithm and O denote the set of packets that OPT sends. Let ADV denote a (modified) adversary. In our proof, we will create ADV and make sure that ADV gains a total value no less than ∑p∈Ovp.
In this paper, we present mdfa, a tool that measures whether a classifier treats differently individuals with similar auditing features but different sensitive attributes. We hope that mdfa’s ability to identify sub-populations with severe violation…
There aren't more papers