Information Bottleneck

Information Bottleneck is a theoretical framework in information theory that formulates the problem of extracting relevant information by compressing a signal while preserving its informativeness about a target variable, characterized by the optimization of a trade-off between compression and prediction accuracy measured via mutual information.

405

Publications

26.9K

Citations

1.2K

Authors

367

Institutions

Publications per year

2017–2026

279

Authors

1.2K

  1. NT
    Naftali Tishby

    Hebrew University of Jerusalem, Israel

    24 publications · 4.6K citations

    20

    h-index

  2. GB
    Gerhard Bauch

    Hamburg University of Technology, Germany

    12 publications · 328 citations

    12

    h-index

  3. XY
    Xiaoqiang Yan

    Zhengzhou University, China

    10 publications · 242 citations

    10

    h-index

  4. NS
    Noam Slonim

    IBM Research - Haifa, Israel

    10 publications · 1.7K citations

    10

    h-index

  5. YY
    Yangdong Ye

    Zhengzhou University, China

    11 publications · 272 citations

    10

    h-index

1–5 of 1.2K

Venues

  1. arXiv (Cornell University)

    Repository · 86 publications · 4.9K citations

    35

    h-index

  2. Entropy

    15 publications · 770 citations

    12

    h-index

  3. Proceedings of the AAAI Conference on Artificial Intelligence

    Conference · 11 publications · 615 citations

    11

    h-index

  4. 10

    h-index

  5. IEEE Transactions on Image Processing

    7 publications · 397 citations

    6

    h-index

1–5