Learning From Play

Learning From Play is a machine learning paradigm within the broader category of Learning From Demonstration that leverages unsupervised, self-generated interaction data from agent exploration—rather than explicit human demonstrations—to infer policies or skills through the identification of structured, reusable behavioral patterns.

188

Publications

14.3K

Citations

517

Authors

206

Institutions

Publications per year

2017–2026

67

Authors

517

  1. MR
    Mark Riedl

    Georgia Institute of Technology, United States

    5 publications · 189 citations

    5

    h-index

  2. MW
    Marco Wiering

    University of Groningen, Netherlands

    3 publications · 112 citations

    3

    h-index

  3. MG
    Matthew Guzdial

    Georgia Institute of Technology, United States

    3 publications · 79 citations

    3

    h-index

  4. SM
    Simon M. Lucas

    University of Essex, United Kingdom

    2 publications · 130 citations

    2

    h-index

  5. DC
    Darrin C. Bentivegna

    Carnegie Mellon University, United States

    2 publications · 177 citations

    2

    h-index

1–5 of 517

Venues

  1. arXiv (Cornell University)

    Repository · 18 publications · 806 citations

    15

    h-index

  2. 5

    h-index

  3. ICGA Journal

    5 publications · 349 citations

    5

    h-index

  4. 3

    h-index

  5. National Conference on Artificial Intelligence

    Conference · 2 publications · 78 citations

    2

    h-index

1–5