The Fundamental Price of Secure Aggregation in Differentially Private Federated Learning Article Swipe

MAKE IMPACT

Weining Chen , Christopher A. Choquette-Choo , Peter Kairouz , Ananda Theertha Suresh · YOU? · · 2022 · Open Access · · DOI: https://doi.org/10.48550/arxiv.2203.03761

We consider the problem of training a $d$ dimensional model with distributed differential privacy (DP) where secure aggregation (SecAgg) is used to ensure that the server only sees the noisy sum of $n$ model updates in every training round. Taking into account the constraints imposed by SecAgg, we characterize the fundamental communication cost required to obtain the best accuracy achievable under $\varepsilon$ central DP (i.e. under a fully trusted server and no communication constraints). Our results show that $\tilde{O}\left( \min(n^2\varepsilon^2, d) \right)$ bits per client are both sufficient and necessary, and this fundamental limit can be achieved by a linear scheme based on sparse random projections. This provides a significant improvement relative to state-of-the-art SecAgg distributed DP schemes which use $\tilde{O}(d\log(d/\varepsilon^2))$ bits per client. Empirically, we evaluate our proposed scheme on real-world federated learning tasks. We find that our theoretical analysis is well matched in practice. In particular, we show that we can reduce the communication cost significantly to under $1.2$ bits per parameter in realistic privacy settings without decreasing test-time performance. Our work hence theoretically and empirically specifies the fundamental price of using SecAgg.

Type

preprint

Language

en

Landing Page

http://arxiv.org/abs/2203.03761

PDF

https://arxiv.org/pdf/2203.03761

OA Status

green

Cited By

11

Related Works

10

OpenAlex ID

https://openalex.org/W4221164659

Raw OpenAlex JSON

OpenAlex ID

https://openalex.org/W4221164659

Canonical identifier for this work in OpenAlex

DOI

https://doi.org/10.48550/arxiv.2203.03761

Digital Object Identifier

Title

The Fundamental Price of Secure Aggregation in Differentially Private Federated Learning

Work title

Type

preprint

OpenAlex work type

Language

en

Primary language

Publication year

2022

Year of publication

Publication date

2022-03-07

Full publication date if available

Authors

Weining Chen, Christopher A. Choquette-Choo, Peter Kairouz, Ananda Theertha Suresh

List of authors in order

Landing page

https://arxiv.org/abs/2203.03761

Publisher landing page

PDF URL

https://arxiv.org/pdf/2203.03761

Direct link to full text PDF

Open access

Yes

Whether a free full text is available

OA status

green

Open access status per OpenAlex

OA URL

https://arxiv.org/pdf/2203.03761

Direct OA link when available

Concepts

Scheme (mathematics), Differential privacy, Federated learning, Computer science, Limit (mathematics), State (computer science), Tilde, Differential (mechanical device), Work (physics), Private information retrieval, Theoretical computer science, Algorithm, Discrete mathematics, Mathematics, Distributed computing, Computer security, Physics, Mathematical analysis, Thermodynamics

Top concepts (fields/topics) attached by OpenAlex

Cited by

11

Total citation count in OpenAlex

Citations by year (recent)

2024: 1, 2023: 9, 2022: 1

Per-year citation counts (last 5 years)

Related works (count)

10

Other works algorithmically related by OpenAlex

Full payload