Locally Differentially Private Sparse Vector Aggregation

@article{Zhou2021LocallyDP,
  title={Locally Differentially Private Sparse Vector Aggregation},
  author={Mingxun Zhou and Tianhao Wang and T-H. Hubert Chan and Giulia C. Fanti and Elaine Shi},
  journal={2022 IEEE Symposium on Security and Privacy (SP)},
  year={2021},
  pages={422-439}
}
Vector mean estimation is a central primitive in federated analytics. In vector mean estimation, each user $i \in[n]$ holds a real-valued vector $v_{i} \in[-1,1]^{d}$, and a server wants to estimate the mean of all n vectors; we would additionally like to protect each user’s privacy. In this paper, we consider the k-sparse version of the vector mean estimation problem. That is, suppose each user’s vector has at most k non-zero coordinates in its d-dimensional vector, and moreover, $k \ll d$. In… 

Figures and Tables from this paper

Improved Utility Analysis of Private CountSketch

This paper considers the classical CountSketch, made differentially private with the Gaussian mechanism, and gives an improved analysis of its estimation error, finding the privacy-utility trade-off is essentially the best one could hope for.

Frequency Estimation in the Shuffle Model with Almost a Single Message

By combining the frequency estimation and the heavy hitter detection protocols, this work shows how to solve the B-dimensional 1-sparse vector summation problem in the high-dimensional setting B=Ω(n), achieving the optimal central-DP MSE Õ(n) with 1 + o(1) messages per user.

Randomize the Future: Asymptotically Optimal Locally Private Frequency Estimation Protocol for Longitudinal Data

The key breakthrough is a new randomizer for sequential data, FutureRand, with two key features: a composition strategy that correlates the noise across the non-zero elements of the sequence, and a pre-computation technique which enables the randomizer to output the results on the fly, without knowing future inputs.

Network change point localisation under local differential privacy

This paper investigates the fundamental limits in consistently localising change points under both node and edge privacy constraints, demon-strating interesting phase transition in terms of the signal-to-noise ratio condition, accompanied by polynomial-time algorithms.

MinMax Sampling: A Near-optimal Global Summary for Aggregation in the Wide Area

This paper proposes MinMax Sampling, a fast, adaptive, and accurate communication scheme for global aggregation in WAN, and designs a scheme, namely MinMaxopt, which trades little accuracy for the other two requirements.

Huff-DP: Huffman Coding based Differential Privacy Mechanism for Real-Time Data

A novel Huffman coding based differential privacy budget selection mechanism (Huff-DP), which selects the optimal privacy budget on the basis of privacy requirement for that specific record, and proposes static, sine, and fuzzy logic based decision algorithms.

References

SHOWING 1-10 OF 58 REFERENCES

Local, Private, Efficient Protocols for Succinct Histograms

Efficient protocols and matching accuracy lower bounds for frequency estimation in the local model for differential privacy are given and it is shown that each user need only send 1 bit to the server in a model with public coins.

Differentially Private Sparse Vectors with Low Error, Optimal Space, and Fast Access

The Approximate Laplace Projection (ALP) mechanism for approximating k-sparse vectors is shown to simultaneously have information-theoretically optimal space, fast access to vector entries, and error of the same magnitude as the Laplace-mechanism applied to dense vectors.

Hadamard Response: Estimating Distributions Privately, Efficiently, and with Little Communication

Hadamard Response (HR) is proposed, a local privatization scheme that requires no shared randomness and is symmetric with respect to the users, and which runs about 100x faster than Randomized Response, RAPPOR, and subset-selection mechanisms.

Heavy Hitter Estimation over Set-Valued Data with Local Differential Privacy

The main idea is to first gather a candidate set of heavy hitters using a portion of the privacy budget, and focus the remaining budget on refining the candidate set in a second phase, which is much more efficient budget-wise than obtaining the heavy hitters directly from the whole dataset.

Communication Complexity in Locally Private Distribution Estimation and Heavy Hitters

This work proposes a sample-optimal $\varepsilon$-locally differentially private (LDP) scheme for distribution estimation, where each user communicates only one bit, and requires no public randomness.

Practical Secure Aggregation for Privacy-Preserving Machine Learning

This protocol allows a server to compute the sum of large, user-held data vectors from mobile devices in a secure manner, and can be used, for example, in a federated learning setting, to aggregate user-provided model updates for a deep neural network.

Prio: Private, Robust, and Scalable Computation of Aggregate Statistics

Pozo is presented, a privacy-preserving system for the collection of aggregate statistics that uses secret-shared non-interactive proofs (SNIPs), a new cryptographic technique that yields a hundred-fold performance improvement over conventional zero-knowledge approaches.

Locally Differentially Private Protocols for Frequency Estimation

This paper introduces a framework that generalizes several LDP protocols proposed in the literature and yields a simple and fast aggregation algorithm, whose accuracy can be precisely analyzed, resulting in two new protocols that provide better utility than protocols previously proposed.

Differentially private summaries for sparse data

This work proposes a general framework for computing the summary directly from the input data, without materializing the vast noisy data, and shows that this is a highly practical solution, which releases a compact summary of the noisy data.

Heavy Hitters and the Structure of Local Privacy

We present a new locally differentially private algorithm for the heavy hitters problem which achieves optimal worst-case error as a function of all standardly considered parameters. Prior work
...