Gradient Leakage Attacks

Gradient Leakage Attacks is a type of privacy attack in federated learning where an adversary reconstructs sensitive raw data or infers private information by analyzing shared model gradients, exploiting the inherent information leakage in gradient updates during distributed training.

78

Publications

5.8K

Citations

295

Authors

99

Institutions

Publications per year

2017–2026

78

Variants

Gradient Inversion Attacks

Authors

295

  1. WW
    Wenqi Wei

    Georgia Institute of Technology, United States

    4 publications · 305 citations

    4

    h-index

  2. LL
    Ling Liu

    Georgia Institute of Technology, United States

    4 publications · 305 citations

    4

    h-index

  3. JC
    Jinyin Chen

    Zhejiang University of Technology, China

    3 publications · 96 citations

    3

    h-index

  4. DI
    Dimitar I. Dimitrov

    ETH Zurich, Switzerland

    3 publications · 44 citations

    3

    h-index

  5. YW
    Yanzhao Wu

    Georgia Institute of Technology, United States

    3 publications · 229 citations

    3

    h-index

1–5 of 295

Venues

  1. arXiv (Cornell University)

    Repository · 18 publications · 1.5K citations

    13

    h-index

  2. 6

    h-index

  3. IEEE Transactions on Big Data

    2 publications · 88 citations

    2

    h-index

  4. 2

    h-index

  5. IEEE/ACM Transactions on Networking

    TON · 2 publications · 84 citations

    2

    h-index

1–5