SOURCE-LINKED INTELLIGENCE

Predicting Privacy Leakage from Weight Spectral Density

arXiv · AI, language, vision and robotics · article · Sep 10, 2026 · UTC

Membership inference attacks (MIAs) are widely used to audit the privacy disclosure risk of machine learning models, however current state-of-the-art attacks require training computationally expensive shadow models, making large-scale privacy evaluation impractical. In this work, we investigate whether inexpensive spectral metrics derived from the heavy-tailed self-regularisation framework can serve as proxies for MIA vulnerability. We evaluate several WeightWatcher spectral metrics on image and tabular classification tasks and compare their relationship with MIA privacy leakage against conven

Read original source ↗ Open in workspace

recordType

paper

region

Global