Reconstruction Attacks

Reconstruction Attacks is a privacy threat in federated learning whereby an adversary infers sensitive private data of individual clients by exploiting shared model updates to reconstruct approximate input samples that were used during local training.

123

Publications

7.3K

Citations

494

Authors

193

Institutions

Publications per year

2017–2026

113

Variants

Data Reconstruction Attacks

Authors

494

  1. AD
    Adam Dziedzic

    University of Chicago, United States

    3 publications · 164 citations

    3

    h-index

  2. RS
    Roei Schuster

    Tel Aviv University, Israel

    3 publications · 164 citations

    3

    h-index

  3. AS
    Adam Smith

    Pennsylvania State University, United States

    3 publications · 390 citations

    3

    h-index

  4. FB
    Franziska Boenisch

    University of Toronto, Canada

    3 publications · 164 citations

    3

    h-index

  5. YC
    Yiran Chen

    Duke University, United States

    3 publications · 81 citations

    3

    h-index

1–5 of 494

Venues

  1. arXiv (Cornell University)

    Repository · 25 publications · 781 citations

    17

    h-index

  2. 10

    h-index

  3. IEEE Transactions on Dependable and Secure Computing

    TDSC · 4 publications · 140 citations

    4

    h-index

  4. IEEE Transactions on Big Data

    3 publications · 123 citations

    3

    h-index

  5. Applied Sciences

    2 publications · 163 citations

    2

    h-index

1–5