🤖 AI Summary
This work investigates the robustness of Local Differential Privacy (LDP) graph analysis under data poisoning attacks, revealing that existing LDP degree estimation protocols severely degrade in accuracy when malicious users inject adversarial noise responses—and that the LDP mechanism itself can amplify the impact of such poisoning. To address this, we propose the first redundancy-driven robust degree estimation protocol, leveraging the intrinsic structural redundancy of graph edges: each edge is independently observed by both incident nodes. Our method integrates robust statistical estimation with strict LDP constraints. Extensive experiments on multiple real-world graph datasets demonstrate that the proposed approach maintains high estimation accuracy even under strong poisoning attacks, reducing mean absolute error by up to 42.6% compared to state-of-the-art baselines. These results empirically validate structural redundancy as an effective and practical source of robustness for LDP graph analytics.
📝 Abstract
Locally differentially private (LDP) graph analysis allows private analysis on a graph that is distributed across multiple users. However, such computations are vulnerable to data poisoning attacks where an adversary can skew the results by submitting malformed data. In this paper, we formally study the impact of poisoning attacks for graph degree estimation protocols under LDP. We make two key technical contributions. First, we observe LDP makes a protocol more vulnerable to poisoning -- the impact of poisoning is worse when the adversary can directly poison their (noisy) responses, rather than their input data. Second, we observe that graph data is naturally redundant -- every edge is shared between two users. Leveraging this data redundancy, we design robust degree estimation protocols under LDP that can significantly reduce the impact of data poisoning and compute degree estimates with high accuracy. We evaluate our proposed robust degree estimation protocols under poisoning attacks on real-world datasets to demonstrate their efficacy in practice.
