Text

A(DP)22SGD : Asynchronous Decentralized Parallel Stochastic Gradient Descent With Differential Privacy

As deep learning models are usually massive and complex, distributed learning is essential for increasing training efficiency. Moreover, in many real-world application scenarios like healthcare, distributed learning can also keep the data local and protect privacy. Recently, the asynchronous decentralized parallel stochastic gradient descent (ADPSGD) algorithm has been proposed and demonstrated to be an efficient and practical strategy where there is no central server, so that each computing node only communicates with its neighbors. Although no raw data will be transmitted across different local nodes, there is still a risk of information leak during the communication process for malicious participants to make attacks. In this paper, we present a differentially private version of asynchronous decentralized parallel SGD framework, or A(DP) 2 SGD for short, which maintains communication efficiency of ADPSGD and prevents the inference from malicious participants. Specifically, Rényi differential privacy is used to provide tighter privacy analysis for our composite Gaussian mechanisms while the convergence rate is consistent with the non-private version. Theoretical analysis shows A(DP) 2 SGD also converges at the optimal O(1/T−−√) rate as SGD. Empirically, A(DP) 2 SGD achieves comparable model accuracy as the differentially private version of Synchronous SGD (SSGD) but runs much faster than SSGD in heterogeneous computing environments.

Ketersediaan

Informasi Detail

Judul Seri

IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE; Vol.44 No.11 Part 2 November 2022

No. Panggil

-

Penerbit

: .,

Deskripsi Fisik

p. 8036-8047

Bahasa

English

ISBN/ISSN

-

Klasifikasi

NONE

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

ASYNCHRONOUS
DISTRIBUTED LEARNING
DIFFERENTIAL PRIVACY
DECENTRALIZED

Info Detail Spesifik

DOI: 10.1109/TPAMI.2021.3107796

Pernyataan Tanggungjawab

Jie Xu, Wei Zhang, Fei Wang

Versi lain/terkait

Lampiran Berkas

Komentar