# statistical

Membership Inference Attacks

Techniques for determining whether specific data was used to train an AI model, including shadow model approaches, loss-based inference, LiRA, and practical implementation guidance.