A privacy-preserving decentralized randomized block-coordinate subgradient algorithm over time-varying networks

作者：

Highlights：

• This paper proposes a privacy-preserving decentralized randomized block-coordinate subgradient projection algorithm.

• It proves that the proposed algorithm is asymptotically convergent.

• It proves that the proposed algorithm can protect the privacy of data.

• It proves that the rate of convergence is achieved, i.e., O(logK/K) and O(logK/K).

摘要

•This paper proposes a privacy-preserving decentralized randomized block-coordinate subgradient projection algorithm.•It proves that the proposed algorithm is asymptotically convergent.•It proves that the proposed algorithm can protect the privacy of data.•It proves that the rate of convergence is achieved, i.e., O(logK/K) and O(logK/K).

论文关键词：Convergence rate,Privacy-preserving,Randomized block-coordinate descent,Subgradient projection

论文评审过程：Received 26 January 2022, Revised 2 July 2022, Accepted 5 July 2022, Available online 11 July 2022, Version of Record 15 July 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118099