LGAICRMay 12, 2021

Accuracy-Privacy Trade-off in Deep Ensemble: A Membership Inference Perspective

arXiv:2105.05381v422 citations
Originality Incremental advance
AI Analysis

This addresses privacy risks in machine learning for practitioners using ensemble methods, highlighting a critical trade-off and offering a solution.

The paper demonstrates an accuracy-privacy trade-off in deep ensembles, showing that membership inference attack effectiveness increases as ensemble accuracy improves across various datasets and architectures, and proposes a defense to simultaneously improve both.

Deep ensemble learning has been shown to improve accuracy by training multiple neural networks and averaging their outputs. Ensemble learning has also been suggested to defend against membership inference attacks that undermine privacy. In this paper, we empirically demonstrate a trade-off between these two goals, namely accuracy and privacy (in terms of membership inference attacks), in deep ensembles. Using a wide range of datasets and model architectures, we show that the effectiveness of membership inference attacks increases when ensembling improves accuracy. We analyze the impact of various factors in deep ensembles and demonstrate the root cause of the trade-off. Then, we evaluate common defenses against membership inference attacks based on regularization and differential privacy. We show that while these defenses can mitigate the effectiveness of membership inference attacks, they simultaneously degrade ensemble accuracy. We illustrate similar trade-off in more advanced and state-of-the-art ensembling techniques, such as snapshot ensembles and diversified ensemble networks. Finally, we propose a simple yet effective defense for deep ensembles to break the trade-off and, consequently, improve the accuracy and privacy, simultaneously.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes