Maximum Mutual Information

Maximum Mutual Information is a model selection criterion that belongs to the family of information-theoretic methods, distinguished by its principle of maximizing the mutual information between input features and latent representations to achieve optimal predictive relevance.

544

Publications

30.1K

Citations

1.4K

Authors

539

Institutions

Publications per year

2017–2026

180

Variants

Mmi

Authors

1.4K

  1. DP
    Daniel Povey

    Johns Hopkins University, United States

    11 publications · 705 citations

    11

    h-index

  2. MV
    Michel Verleysen

    UCLouvain, Belgium

    10 publications · 780 citations

    10

    h-index

  3. PC
    Philip C. Woodland

    University of Cambridge, United Kingdom

    9 publications · 379 citations

    9

    h-index

  4. YN
    Yves Normandin

    Computer Research Institute of Montréal, Canada

    7 publications · 403 citations

    7

    h-index

  5. DE
    Deniz Erdoğmuş

    Northeastern University, United States

    6 publications · 254 citations

    6

    h-index

1–5 of 1.4K

Venues

  1. Neurocomputing

    20 publications · 1.2K citations

    19

    h-index

  2. arXiv (Cornell University)

    Repository · 25 publications · 1.3K citations

    17

    h-index

  3. IEEE Transactions on Information Theory

    19 publications · 806 citations

    16

    h-index

  4. Pattern Recognition

    14 publications · 1.4K citations

    13

    h-index

  5. Expert Systems with Applications

    10 publications · 830 citations

    10

    h-index

1–5