← all papers · overview

Adversarial Sample-Based Approach for Tighter Privacy Auditing in Final Model-Only Scenarios

Abstract

Auditing Differentially Private Stochastic Gradient Descent (DP-SGD) in the final model setting is challenging and often results in empirical lower bounds that are significantly looser than theoretical privacy guarantees. We introduce a novel auditing method that achieves tighter empirical lower bounds without additional assumptions by crafting worst-case adversarial samples through loss-based input-space auditing. Our approach surpasses traditional canary-based heuristics and is effective in final model-only scenarios. Specifically, with a theoretical privacy budget of , our method achieves empirical lower bounds of , compared to the baseline of for MNIST. Our work offers a practical framework for reliable and accurate privacy auditing in differentially private machine learning.

Related papers

Ranked by semantic similarity — how closely each paper's abstract matches this one (100% = near-identical topic).