The UTS AAII Research Seminar Series – Seminar 13

Secure inference is designed to enable encrypted machine learning model prediction over encrypted data. It will ease privacy concerns when models are deployed in Machine Learning as a Service.

For efficiency, most of recent secure inference protocols are constructed using secure multi-party computation (MPC) techniques. However, MPC-based protocols do not hide information revealed from their output. In the context of secure inference, prediction outputs (i.e., inference results of encrypted user inputs and models) are revealed to the users.

As a result, adversaries can compromise output privacy of secure inference, i.e., launching Membership Inference Attacks (MIAs) by querying encrypted models, just like MIAs in plaintext inference.

In this talk, I will first share our observations on the vulnerability of MPC-based secure inference to MIAs, though it yields perturbed predictions due to approximations. Then I will report on our recent research effort in guarding the output privacy of secure inference from being exploited by MIAs. I will also discuss the future research along with the line of privacy-preserving machine learning and deep learning.