Semi-supervised Active Learning

Semi-Supervised Active Learning is an active learning paradigm that integrates a small set of initially labeled data with a large pool of unlabeled data, distinguishing itself from conventional active learning by leveraging both supervised learning on labeled instances and semi-supervised learning on unlabeled data to iteratively select the most informative samples for labeling, thereby improving model performance with reduced annotation effort.

271

Publications

18K

Citations

1K

Authors

356

Institutions

Publications per year

2017–2026

134

Authors

1K

  1. MR
    Michael R. Lyu

    Chinese University of Hong Kong, Hong Kong SAR China

    4 publications · 460 citations

    4

    h-index

  2. SC
    Steven C. H. Hoi

    Nanyang Technological University, Singapore

    4 publications · 460 citations

    4

    h-index

  3. XZ
    Xiaojin Zhu

    University of Wisconsin–Madison, United States

    4 publications · 1.2K citations

    4

    h-index

  4. JL
    Jun Li

    Sun Yat-sen University, China

    4 publications · 573 citations

    4

    h-index

  5. ZZ
    Zhi‐Hua Zhou

    Nanjing University, China

    4 publications · 433 citations

    4

    h-index

1–5 of 1K

Venues

  1. arXiv (Cornell University)

    Repository · 16 publications · 958 citations

    13

    h-index

  2. 7

    h-index

  3. Neurocomputing

    5 publications · 284 citations

    5

    h-index

  4. Information Sciences

    5 publications · 238 citations

    5

    h-index

  5. PLoS ONE

    4 publications · 140 citations

    4

    h-index

1–5