Students Parrot Their Teachers: Membership Inference on Model Distillation
This work addresses privacy concerns for machine learning practitioners using distillation, revealing that it is an incremental improvement with significant vulnerabilities.
The paper tackles the problem of privacy leakage in model distillation by designing membership inference attacks, showing that distillation alone provides only limited privacy across multiple domains, with attacks being strongest when student and teacher sets are similar or when the teacher set is poisoned.
Model distillation is frequently proposed as a technique to reduce the privacy leakage of machine learning. These empirical privacy defenses rely on the intuition that distilled ``student'' models protect the privacy of training data, as they only interact with this data indirectly through a ``teacher'' model. In this work, we design membership inference attacks to systematically study the privacy provided by knowledge distillation to both the teacher and student training sets. Our new attacks show that distillation alone provides only limited privacy across a number of domains. We explain the success of our attacks on distillation by showing that membership inference attacks on a private dataset can succeed even if the target model is *never* queried on any actual training points, but only on inputs whose predictions are highly influenced by training data. Finally, we show that our attacks are strongest when student and teacher sets are similar, or when the attacker can poison the teacher set.