Differentially Private Analysis of Graphs

Raskhodnikova, Sofya; Smith, Adam

doi:10.1007/978-1-4939-2864-4_549

Cited by 14 publications

(12 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Here, we assume that the server and other users can be honest-butcurious adversaries and can obtain all edges in G other than edges of v i as background knowledge. To strongly protect edges of v i from these adversaries, we use edge DP [51], [52] as a privacy metric. Specifically, edge DP hides one edge between any two users from the adversary.…”

Section: B Local Differential Privacy On Graphsmentioning

confidence: 99%

Degree-Preserving Randomized Response for Graph Neural Networks under Local Differential Privacy

Hidano¹,

Murakami²

2022

Preprint

View full text Add to dashboard Cite

Differentially private GNNs (Graph Neural Networks) have been recently studied to provide high accuracy in various tasks on graph data while strongly protecting user privacy. In particular, a recent study proposes an algorithm to protect each user's feature vector in an attributed graph with LDP (Local Differential Privacy), a strong privacy notion without a trusted third party. However, this algorithm does not protect edges (friendships) in a social graph or protect user privacy in unattributed graphs. It remains open how to strongly protect edges with LDP while keeping high accuracy in GNNs.In this paper, we propose a novel LDP algorithm called the DPRR (Degree-Preserving Randomized Response) to provide LDP for edges in GNNs. Our DPRR preserves each user's degree hence a graph structure while providing edge LDP. Technically, we use Warner's RR (Randomized Response) and strategic edge sampling, where each user's sampling probability is automatically tuned to preserve the degree information. We prove that the DPRR approximately preserves the degree information under edge LDP. We focus on graph classification as a task of GNNs and evaluate the DPRR using two social graph datasets. Our experimental results show that the DPRR significantly outperforms Warner's RR and provides accuracy close to a non-private algorithm with a reasonable privacy budget, e.g., ε = 1.

show abstract

Section: B Local Differential Privacy On Graphsmentioning

confidence: 99%

Degree-Preserving Randomized Response for Graph Neural Networks under Local Differential Privacy

Hidano¹,

Murakami²

2022

Preprint

View full text Add to dashboard Cite

show abstract

“…DP on Graphs. For graphs, we can consider two types of DP: edge DP and node DP [29,57]. Edge DP hides the existence of one edge, whereas node DP hides the existence of one node along with its adjacent edges.…”

Section: Differential Privacymentioning

confidence: 99%

Differentially Private Subgraph Counting in the Shuffle Model

Imola¹,

Murakami²,

Chaudhuri³

2022

Preprint

View full text Add to dashboard Cite

Subgraph counting is fundamental for analyzing connection patterns or clustering tendencies in graph data. Recent studies have applied LDP (Local Differential Privacy) to subgraph counting to protect user privacy even against a data collector in both centralized and decentralized social networks. However, existing local algorithms suffer from extremely large estimation errors or assume multi-round interaction between users and the data collector, which requires a lot of user effort and synchronization.In this paper, we focus on a one-round of interaction and propose accurate subgraph counting algorithms by introducing a recently studied shuffle model. We first propose a basic technique called wedge shuffling to send wedge information, the main component of several subgraphs, with small noise. Then we apply our wedge shuffling to counting triangles and 4-cycles -basic subgraphs for analyzing clustering tendencies -with several additional techniques. We also show upper bounds on the estimation error for each algorithm. We show through comprehensive experiments that our one-round shuffle algorithms significantly outperform the one-round local algorithms in terms of accuracy and achieve small estimation errors with a reasonable privacy budget, e.g., smaller than 1 in edge DP.

show abstract

“…When we apply LDP to graphs, we follow the direction of edge DP [43,48] that has been developed for the central DP model. In edge DP, the existence of an edge between any two users is protected; i.e., two computations, one using a graph with the edge and one using the graph without the edge, are indistinguishable.…”

Section: Local Differential Privacy On Graphsmentioning

confidence: 99%

Communication-Efficient Triangle Counting under Local Differential Privacy

Imola¹,

Murakami²,

Chaudhuri³

2021

Preprint

View full text Add to dashboard Cite

Triangle counting in networks under LDP (Local Differential Privacy) is a fundamental task for analyzing connection patterns or calculating a clustering coefficient while strongly protecting sensitive friendships from a central server. In particular, a recent study proposes an algorithm for this task that uses two rounds of interaction between users and the server to significantly reduce estimation error. However, this algorithm suffers from a prohibitively large communication cost due to a large noisy graph each user needs to download.In this work, we propose triangle counting algorithms under LDP with a small estimation error and communication cost. We first propose two-round algorithms consisting of edge sampling and carefully selecting edges each user downloads so that the estimation error is small. Then we propose a double clipping technique, which clips the number of edges and then the number of noisy triangles, to significantly reduce the sensitivity of each user's query. Through comprehensive evaluation, we show that our algorithms dramatically reduce the communication cost of the existing algorithm, e.g., from 6 hours to 8 seconds or less at 20 Mbps download rate, while keeping a small estimation error.

show abstract

Differentially Private Analysis of Graphs

Cited by 14 publications

References 9 publications

Degree-Preserving Randomized Response for Graph Neural Networks under Local Differential Privacy

Degree-Preserving Randomized Response for Graph Neural Networks under Local Differential Privacy

Differentially Private Subgraph Counting in the Shuffle Model

Communication-Efficient Triangle Counting under Local Differential Privacy

Contact Info

Product

Resources

About