Skip to content

Reference Model Membership Inference (LBRM)

Home / Concepts / Reference Model Membership Inference (LBRM)

Reference Model Membership Inference (LBRM)

Auto-generated stub. Edit this file to add more details.

A two-stage membership inference attack framework for time series imputation models that utilizes a reference model to improve the detection of training data inclusion.

Why It Matters

It is a novel membership inference attack specifically tailored for time series imputation models, designed to work even when models are robust to traditional overfitting-based attacks.

Evidence

a novel membership inference attack based on a reference model that improves detection accuracy

Metadata & Links