Activation Annealing

Activation Annealing is a computational optimization technique in machine learning that belongs to the class of annealing methods, distinguished by its systematic adjustment of neuron activation states to escape local minima during model training.

27

Publications

802

Citations

156

Authors

49

Institutions

Publications per year

2017–2026

6

Parents

Authors

156

  1. JH
    Jung Han

    Yale University, United States

    2 publications · 78 citations

    2

    h-index

  2. JO
    Jihun Oh

    Korea Advanced Institute of Science and Technology, South Korea

    2 publications · 46 citations

    2

    h-index

  3. JS
    Jun Suda

    Kyoto University, Japan

    2 publications · 70 citations

    2

    h-index

  4. TK
    Tetsu Kachi

    Toyota Central Research and Development Laboratories (Japan), Japan

    2 publications · 59 citations

    2

    h-index

  5. AS
    Akio Shima

    Hitachi (Japan), Japan

    2 publications · 34 citations

    2

    h-index

1–5 of 156

Venues

  1. Applied Physics Letters

    5 publications · 166 citations

    5

    h-index

  2. Japanese Journal of Applied Physics

    JJAP · 3 publications · 42 citations

    3

    h-index

  3. Journal of Applied Physics

    3 publications · 75 citations

    3

    h-index

  4. IEEE Electron Device Letters

    3 publications · 177 citations

    3

    h-index

  5. Semiconductor Science and Technology

    2 publications · 46 citations

    2

    h-index

1–5