Membership inference attacks (MIAs) are a type of privacy attack in which an adversary attempts to infer whether a given data point was used to train a machine learning model. MIAs can... Read more »